| Logo | Name | Excerpt | Parameters | Quantizations | VRAM | License | Last Update | Action |
|---|---|---|---|---|---|---|---|---|
| No Logo | Qwen2.5Vl 3B | "Qwen2.5Vl 3B: Alibaba's 3B param model excels in image & video analysis, offering monolingual support with a context-length of up to 3,2k." | 4B | fp16 q4 q8 | 2.0 GB |
Apache-2.0 |
2025-05-22 | Details |
| No Logo | Qwen2.5Vl 7B | "Qwen2.5VL 7B: Alibaba's 7 billion parameter LLM for visual content analysis." | 8B | fp16 q4 q8 | 4.0 GB |
Apache-2.0 |
2025-05-22 | Details |
| No Logo | Qwen2.5Vl 32B | "Qwen2.5Vl 32B: Alibaba's 32 billion parameter LLM, excelling in visual question answering with a 32k context window." | 34B | fp16 q4 q8 | 17.0 GB |
Apache-2.0 |
2025-05-22 | Details |
| No Logo | Qwen2.5Vl 72B | "Qwen2.5Vl 72B: Alibaba's 72 billion parameter LLM for image & video analysis." | 73B | fp16 q4 q8 | 36.5 GB |
Apache-2.0 |
2025-05-22 | Details |
|
|
Devstral 24B | "Devstral 24B by Mistral AI: A powerful 24 billion parameter LLM for coding tasks. Supports context lengths up to 8k." | 24B | fp16 q4 q8 | 12.0 GB |
Apache-2.0 |
2025-05-21 | Details |
| No Logo | Qwen3 0.6B | "Qwen3 0.6B: Alibaba's 0.6 billion param model excels in reasoning tasks with a 32k context window." | 1B | fp16 q4 q8 | 0.5 GB |
Apache-2.0 |
2025-05-03 | Details |
| No Logo | Qwen3 1.7B | "Qwen3 1.7B: Alibaba's 1.7 billion param model for enhanced reasoning & problem-solving in English." | 2B | fp16 q4 q8 | 1.0 GB |
Apache-2.0 |
2025-05-03 | Details |
| No Logo | Qwen3 4B | "Qwen3 4B: Alibaba's 4 billion parameter LLM for enhanced reasoning & logic tasks." | 4B | fp16 q4 q8 | 2.0 GB |
Apache-2.0 |
2025-05-03 | Details |
| No Logo | Qwen3 8B | "Qwen3 8B: Alibaba's 8 billion parameter LLM for reasoning & problem-solving. Mono-lingual model with context lengths up to 8k." | 8B | fp16 q4 q8 | 4.0 GB |
Apache-2.0 |
2025-05-03 | Details |
| No Logo | Qwen3 14B | "Qwen3 14B: Alibaba's 14 billion parameter LLM for advanced reasoning tasks." | 15B | fp16 q4 q8 | 7.5 GB |
Apache-2.0 |
2025-05-03 | Details |
| No Logo | Qwen3 30B | "Qwen3 30B: Alibaba's 30 billion param large language model for advanced reasoning tasks." | 30B | fp16 q4 q8 | 15.0 GB |
Apache-2.0 |
2025-05-03 | Details |
| No Logo | Qwen3 32B | "Qwen3 32B: Alibaba's 32 billion param LLM for complex question answering with logical reasoning." | 33B | fp16 q4 q8 | 16.5 GB |
Apache-2.0 |
2025-05-03 | Details |
| No Logo | Qwen3 235B | "Qwen3 235B: Alibaba's large language model with 235 billion parameters, designed for text generation tasks." | 235B | fp16 q4 q8 | 117.5 GB |
Apache-2.0 |
2025-05-03 | Details |
|
|
Phi4 Reasoning 14B | "Phi4 Reasoning 14B: Microsoft's 14 billion param model, designed for accelerating language model research and generative AI feature development." | 15B | fp16 q4 q8 | 7.5 GB |
Microsoft |
2025-05-01 | Details |
|
|
Phi4 Mini Reasoning 3.8B | "Phi4 Mini Reasoning 3.8B by Microsoft: A 3.8 billion parameter LLM excelling in multi-step, logic-intensive math tasks." | 4B | fp16 q4 q8 | 2.0 GB |
Microsoft |
2025-05-01 | Details |
| No Logo | Llama4 400B Instruct | "Llama4 400B Instruct by Meta Llama Enterprise: A versatile, multilingual large language model with 400 billion parameters, designed for commercial and research applications." | 402B | fp16 q4 q8 | 201.0 GB |
None None |
2025-04-24 | Details |
| No Logo | Llama4 109B Instruct | "Llama4 109B Instruct by Meta Llama Enterprise: A powerful, multi-lingual large language model with 109 billion parameters, designed for commercial applications." | 109B | fp16 q4 q8 | 54.5 GB |
None None |
2025-04-24 | Details |
|
|
Gemma3 4B Instruct | "Gemma3 4B by Google: 4 billion params, 128k/32k context-length. Supports multiple languages for creative content generation, chatbot AI, text summarization, and image data extraction." | 4B | fp16 q4 q8 | 2.0 GB |
Gemma-Terms-of-Use |
2025-04-18 | Details |
|
|
Gemma3 1B Instruct | "Gemma3 1B by Google: Multilingual LLM with 1B params, 128k/32k context-length. Ideal for content creation & communication tasks." | 1B | fp16 q4 q8 | 0.5 GB |
Gemma-Terms-of-Use |
2025-04-18 | Details |
|
|
Gemma3 12B Instruct | "Gemma3 12B: Google's 12 billion param. LLM for multilingual content creation & comm., incl. text gen., chatbots, summarization & image data extraction." | 12B | fp16 q4 q8 | 6.0 GB |
Gemma-Terms-of-Use |
2025-04-18 | Details |
|
|
Gemma3 27B Instruct | "Gemma3 27B: Google's 27 billion parameter LLM for creative content & comms. Supports context up to 128k tokens. Ideal for text gen, chatbots, summarization & image data extraction." | 27B | fp16 q4 q8 | 13.5 GB |
Gemma-Terms-of-Use |
2025-04-18 | Details |
|
|
Granite3.3 2B | "Granite3.3 2B: IBM's 2 billion param., multi-lingual LLM for instruction-following tasks." | 3B | q4 | 1.5 GB |
Apache-2.0 |
2025-04-17 | Details |
|
|
Granite3.3 8B | "Granite3.3 8B: IBM's 8 billion param., multi-lingual LLM for instruction-following tasks." | 8B | q4 | 4.0 GB |
Apache-2.0 |
2025-04-17 | Details |
|
|
Mistral Small3.1 24B Instruct | "Mistral Small3.1 24B Instruct: Rapid multilingual conversational AI by Mistral AI." | 24B | fp16 q4 q8 | 12.0 GB |
Apache-2.0 |
2025-04-13 | Details |
| No Logo | Cogito 14B | "Cogito 14B: 14 billion param. LLM by Deep Cogito. Supports multilingual coding assistance with context lengths up to 8k." | 15B | fp16 q4 q8 | 7.5 GB |
Apache-2.0 |
2025-04-13 | Details |
| No Logo | Cogito 32B | "Cogito 32B: Deep Cogito's 32 billion param model for coding assistance. Mono-lingual with context lengths up to 8k." | 33B | fp16 q4 q8 | 16.5 GB |
Apache-2.0 |
2025-04-13 | Details |
| No Logo | Cogito 3B | "Cogito 3B: A 3 billion parameter LLM by Deep Cogito, excelling in coding & STEM tasks. Supports multiple context lengths up to 8k." | 4B | fp16 q4 q8 | 2.0 GB |
LLAMA-32-COMMUNITY |
2025-04-13 | Details |
| No Logo | Cogito 70B | "Cogito 70B: Deep Cogito's large language model with 70 billion parameters, designed for coding assistance. Supports multiple context lengths up to 8k." | 71B | fp16 q4 q8 | 35.5 GB |
Apache-2.0 |
2025-04-13 | Details |
| No Logo | Cogito 8B | "Cogito 8B: 8 billion param. LLM by Deep Cogito for coding aid." | 8B | fp16 q4 q8 | 4.0 GB |
LLAMA-32-COMMUNITY |
2025-04-13 | Details |
| No Logo | Deepcoder 1.5B | "Deepcoder 1.5B: Agentica's 1.5 billion param model for coding assistance, with a 64k context window." | 2B | fp16 q4 q8 | 1.0 GB |
MIT |
2025-04-13 | Details |
| No Logo | Deepcoder 14B | "Deepcoder 14B by Agentica: 14 billion param. LLM for coding aid, supports single language." | 15B | fp16 q4 q8 | 7.5 GB |
MIT |
2025-04-13 | Details |
| No Logo | Openthinker 7B | test | 8B | fp16 q4 q8 | 4.0 GB |
Apache-2.0 None |
2025-04-11 | Details |
| No Logo | Openthinker 32B | "Openthinker 32B by Bespoke Labs: A versatile, 32-billion parameter model with a 16k context window, designed for multilingual language understanding and generation tasks." | 33B | fp16 q4 q8 | 16.5 GB |
Apache-2.0 None |
2025-04-11 | Details |
|
|
Mistral Small 22B Instruct | "Mistral Small 22B: A 22 billion parameter LLM by Mistral AI, with a 32k context window, designed for efficient, mono-lingual inference pipeline implementation." | 22B | fp16 q2 q3 q4 q5 q6 q8 | 11.0 GB |
Not Found Not Found |
2025-04-07 | Details |
| No Logo | Mistral Small 24B Instruct | 24B | fp16 q2 q3 q4 q5 q6 q8 | 12.0 GB |
Apache-2.0 |
2025-04-07 | Details | |
| No Logo | Exaone Deep 7.8B | "Exaone Deep 7.8B by LG AI Research: A 7.8 billion parameter LLM with a 32k context window, excelling in mathematical problem-solving." | 8B | fp16 q4 q8 | 4.0 GB |
EXAONE-AIMLA-11-NC |
2025-03-25 | Details |
| No Logo | Exaone Deep 2.4B | "Exaone Deep 2.4B by LG AI Research: A 2.4 billion parameter monolithic model for general text generation." | 3B | fp16 q4 q8 | 1.5 GB |
EXAONE-AIMLA-11-NC |
2025-03-25 | Details |
| No Logo | Exaone Deep 32B | "Exaone Deep 32B by LG AI Research: A powerful 32 billion parameter model for mathematical problem-solving and reasoning." | 32B | fp16 q4 q8 | 16.0 GB |
EXAONE-AIMLA-11-NC |
2025-03-25 | Details |
| No Logo | Qwq 32B | "Qwq 32B: Alibaba's 32 billion param, 32k context-length LLM for efficient math assistance." | 33B | fp16 q4 q8 | 16.5 GB |
Apache-2.0 |
2025-03-16 | Details |
|
|
Command A 111B | "Command A 111B by Cohere AI: Multilingual LLM with 111 billion params, context lengths up to 6k, excelling in conversational AI." | 111B | fp16 q4 q8 | 55.5 GB |
CC-BY-NC-4.0 |
2025-03-16 | Details |
|
|
Granite3.2 8B Instruct | "Granite3.2 8B Instruct: IBM's 8 billion param, multilingual model for versatile instruction following." | 8B | fp16 q4 q8 | 4.0 GB |
Apache-2.0 |
2025-03-01 | Details |
|
|
Granite3.2 2B Instruct | "Granite3.2 2B Instruct: IBM's multilingual, 2 billion parameter model for versatile instruction following." | 3B | fp16 q4 q8 | 1.5 GB |
Apache-2.0 |
2025-03-01 | Details |
|
|
Phi4 Mini 3.8B | "Phi4 Mini 4B: Microsoft's 4 billion param, multilingual LLM with 128k context-length for versatile AI systems." | 4B | fp16 q4 q8 | 2.0 GB |
Microsoft |
2025-03-01 | Details |
|
|
Granite3.2 Vision 2B | "Granite3.2 Vision 3B: IBM's 3 billion param, monolingual LLM for enterprise visual & text data processing." | 3B | fp16 q4 q8 | 1.5 GB |
Apache-2.0 |
2025-03-01 | Details |
|
|
Command R7B Arabic 7B | "Command R7B Arabic 8B: Google's 7B param, bi-lingual model for 128k context-length tasks, excelling in instruction following." | 8B | fp16 q4 q8 | 4.0 GB |
CC-BY-NC-4.0 |
2025-03-01 | Details |
| No Logo | Deepseek R1 8B | "Deepseek R1 8B: Bi-lingual Large Language Model by Lm Studio Community. 8 billion params, supports context lengths up to 128k for efficient code generation & debugging." | 8B | fp16 q4 q8 | 4.0 GB |
MIT |
2025-02-27 | Details |
|
|
Deepseek R1 1.5B | "Deepseek R1 1.5B: Bi-lingual Large Language Model with 1.5 billion parameters, supporting context lengths up to 128k for enhanced code generation and debugging." | 2B | fp16 q4 q8 | 1.0 GB |
MIT |
2025-02-27 | Details |
|
|
Deepseek R1 14B | "Deepseek R1 14B: Bi-lingual Large Language Model with 14 billion parameters. Supports context lengths up to 128k for enhanced code generation and debugging." | 15B | fp16 q4 q8 | 7.5 GB |
MIT |
2025-02-27 | Details |
|
|
Deepseek R1 32B | "Deepseek R1 32B: 32 billion param. LLM for bilingual coding tasks. Supports context lengths up to 128k." | 33B | fp16 q4 q8 | 16.5 GB |
MIT |
2025-02-27 | Details |
|
|
Deepseek R1 70B | "Deepseek R1 70B: Bi-lingual Large Language Model with 70 billion parameters. Ideal for code generation & debugging." | 71B | fp16 q4 q8 | 35.5 GB |
MIT |
2025-02-27 | Details |
|
|
Deepseek R1 671B | "Deepseek R1 671B: Bi-lingual Large Language Model with 671 billion parameters. Supports context lengths up to 128k for advanced code generation and debugging tasks." | 671B | fp16 q4 q8 | 335.5 GB |
MIT |
2025-02-27 | Details |
|
|
Deepscaler 1.5B | "Deepscaler 1.5B by Deepseek: A large language model with 1.5 billion parameters, supporting context lengths up to 24k for enhanced educational applications like solving math problems and comprehending complex concepts." | 2B | fp16 q4 q8 | 1.0 GB |
MIT |
2025-02-27 | Details |
|
|
Deepseek R1 7B | "Deepseek R1 7B: Bi-lingual Large Language Model with 7 billion parameters, excelling in code generation & debugging." | 8B | fp16 q4 q8 | 4.0 GB |
MIT |
2025-02-27 | Details |
| No Logo | R1 1776 70B | "R1 1776 70B by Perplexity Enterprise: A powerful monolithic model with 70 billion parameters, designed for complex question answering and reasoning tasks." | 71B | fp16 q4 q8 | 35.5 GB |
MIT |
2025-02-22 | Details |
| No Logo | R1 1776 671B | "R1 1776 671B: Perplexity's 671B param model for complex, reasoned answers in English." | 671B | fp16 q4 q8 | 335.5 GB |
MIT |
2025-02-22 | Details |
|
|
Granite3.1 Dense 8B Instruct | "Granite3.1 Dense 8B: IBM's 8 billion param, multilingual LLM for efficient text summarization." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 |
2025-01-27 | Details |
|
|
Granite3.1 Dense 2B Instruct | "Granite3.1 Dense 2B: IBM's multilingual, 2 billion param model for efficient text summarization with a 128k context window." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Apache-2.0 |
2025-01-27 | Details |
|
|
Granite3.1 Moe 3B Instruct | "Granite3.1 Moe 3B: IBM's 3 billion param, multilingual LLM for summarization, with a context-length of 128k." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Apache-2.0 |
2025-01-27 | Details |
|
|
Granite3.1 Moe 1B Instruct | "Granite3.1 Moe 1B: IBM's 1B param, multilingual LLM for efficient text summarization." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
Apache-2.0 |
2025-01-27 | Details |
| No Logo | Smallthinker 3B | "Smallthinker 3B: A compact, 3 billion parameter LLM by Powerinfer. Ideal for edge deployment with a 16k context window and monolingual language support." | 3B | fp16 q4 q8 | 1.5 GB |
Qwen-RESEARCH |
2025-01-27 | Details |
|
|
Dolphin3 8B | 8B | fp16 q4 q8 | 4.0 GB |
LLAMA-31-CCLA |
2025-01-27 | Details | |
|
|
Phi4 14B | "Phi4 15B: Microsoft's 15 billion parameter, monolingual language model with a 16k context window, accelerating linguistic research." | 15B | fp16 q4 q8 | 7.5 GB |
Microsoft |
2025-01-27 | Details |
|
|
Command R7B 7B | "Command R7B 8B by Cohere AI: 7B param, 128k context, multilingual, excels in reasoning tasks." | 8B | fp16 q4 q8 | 4.0 GB |
CC-BY-NC-4.0 |
2025-01-27 | Details |
| No Logo | Olmo2 7B Instruct | "Olmo2 7B: AI2's 7 billion param, 4k context-length model for monolingual text tasks." | 7B | fp16 q4 q8 | 3.5 GB |
Apache-2.0 |
2025-01-27 | Details |
| No Logo | Olmo2 13B Instruct | "Olmo2 13B: AI2's 13 billion param, 4k context-length model for research. Mono-lingual." | 14B | fp16 q4 q8 | 7.0 GB |
Apache-2.0 |
2025-01-27 | Details |
|
|
Deepseek V3 671B | "Deepseek V3 671B: Bi-lingual, 128k context-length LLM for NLP tasks." | 671B | fp16 q4 q8 | 335.5 GB |
DEEPSEEK-LICENSE |
2025-01-27 | Details |
|
|
Granite Embedding 278M | "Granite Embedding 278M by IBM Granite: A versatile, multi-lingual model with 278M parameters, optimized for text similarity tasks." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-12-31 | Details |
|
|
Granite Embedding 30M | "Granite Embedding 30M by Lm Studio Community: A 30M parameter, monolingual model optimized for text similarity tasks." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-12-31 | Details |
|
|
Hermes3 8B | "Hermes3 8B: An 8 billion parameter, mono-lingual large language model by Nousresearch, designed for general assistance tasks." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-12-31 | Details |
|
|
Hermes3 70B | "Hermes3 70B: NousResearch's large language model with 70 billion parameters, a context window of 4k, and monolingual support. Ideal for general assistant tasks." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-12-31 | Details |
| No Logo | Hermes3 405B | "Hermes3 405B: Open-source LLM by Techhub, 405B params, 4k context, monolingual." | 406B | fp16 q2 q3 q4 q5 q6 q8 | 203.0 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-12-31 | Details |
|
|
Mixtral 8X7B | Mixtral 8x7B by Mistral AI: 7 billion param, 2k context, mono-lingual, excels in text generation. | 47B | fp16 q2 q3 q4 q5 q6 q8 | 23.5 GB |
Apache-2.0 Apache-2.0 |
2024-12-31 | Details |
|
|
Mixtral 8X22B | Mixtral 8x22B by Mistral AI: 22B param, 2k context, mono-lingual, excels in text generation. | 141B | fp16 q2 q3 q4 q5 q6 q8 | 70.5 GB |
Apache-2.0 Apache-2.0 |
2024-12-31 | Details |
|
|
Mixtral 8X7B Instruct | Mixtral 8x7B Instruct, maintained by Mistral AI, is an open-source large language model with 7 billion parameters, designed to follow instructions and engage in conversations. | 47B | fp16 q2 q3 q4 q5 q6 q8 | 23.5 GB |
Apache-2.0 Apache-2.0 |
2024-12-31 | Details |
|
|
Mixtral 8X22B Instruct | Mixtral 8x22B Instruct by Mistral AI: 22B param, 64k context, mono-lingual, excels in text generation. | 141B | fp16 q2 q3 q4 q5 q6 q8 | 70.5 GB |
Apache-2.0 Apache-2.0 |
2024-12-31 | Details |
|
|
Dolphin Mixtral 8X7B | "Dolphin 8X7B: A 7 billion parameter LLM by Cognitive Computations. Supports up to 32k context length for efficient coding assistance." | 47B | fp16 q2 q3 q4 q5 q6 q8 | 23.5 GB |
Apache-2.0 Apache-2.0 |
2024-12-31 | Details |
|
|
Dolphin Mixtral 8X22B | "Dolphin Mixtral 8X22B: A powerful 22 billion parameter model by Cognitive Computations, excelling in instructional tasks with a vast context-length of up to 64k tokens." | 141B | fp16 q2 q3 q4 q5 q6 q8 | 70.5 GB |
Apache-2.0 Apache-2.0 |
2024-12-31 | Details |
| No Logo | Tulu3 8B | "Tulu3 8B by AI2 Enterprise: An 8 billion parameter, monolingual model for content generation with a context-length of 2k." | 8B | fp16 q4 q8 | 4.0 GB |
LLAMA-31-CCLA |
2024-12-31 | Details |
| No Logo | Marco O1 7B | "Marco O1 8B: Aidc-Ai's 8 billion param, bi-lingual LLM with 2k context-length for open-ended tasks." | 8B | fp16 q4 q8 | 4.0 GB |
Apache-2.0 |
2024-12-31 | Details |
| No Logo | Sailor2 1B | "Sailor2 1B: A versatile, 1 billion parameter large language model by Sea AI Lab. Supports multilingual contexts up to 2048 tokens for efficient production use." | 1B | fp16 q4 q8 | 0.5 GB |
Apache-2.0 |
2024-12-31 | Details |
|
|
Snowflake Arctic Embed2 568M | "Snowflake's Arctic Embed2 1B: A multilingual, 1 billion parameter LLM with an 8k context window, optimized for text retrieval & search." | 1B | fp16 | 0.5 GB | 2024-12-31 | Details | |
| No Logo | Llama3.3 70B Instruct | "Llama3.3 70B: Meta's 70 billion param, multi-lingual model for assistant-style chat, with a context window of 128k." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
Llama-33-AUP LLAMA-33-CCLA |
2024-12-31 | Details |
| No Logo | Tulu3 70B | "Tulu3 70B by AI2 Enterprise: 70 billion param. LLM for research & education, supports English with context lengths up to 8K." | 71B | fp16 q4 q8 | 35.5 GB |
LLAMA-31-CCLA |
2024-12-31 | Details |
| No Logo | Exaone3.5 32B Instruct | "Exaone3.5 32B: LG AI's 32 billion param, 32k context-length LLM for bilingual conversation & assistance." | 32B | fp16 q4 q8 | 16.0 GB |
EXAONE-AIMLA-11-NC |
2024-12-31 | Details |
|
|
Hermes3 3B | "Hermes3 3B: NousResearch's open-source large language model with 3 billion parameters, supporting context lengths up to 32K for versatile, monolingual applications." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
META-LLAMA-3-CCLA |
2024-12-31 | Details |
|
|
Falcon3 7B Instruct | "Falcon3 7B: TII's 7 billion param, multilingual model for extensive text generation with a 32k context window." | 7B | fp16 q4 q8 | 3.5 GB |
Falcon-3-TII |
2024-12-31 | Details |
|
|
Falcon3 1B Instruct | "Falcon3 1B: TII's 1 billion param, multilingual LLM. Excels in complex reasoning tasks with a 4k context window." | 2B | fp16 q4 q8 | 1.0 GB |
Falcon-3-TII |
2024-12-31 | Details |
|
|
Falcon3 3B Instruct | "Falcon3 3B: TII's multilingual, 8k context-length model for advanced text generation." | 3B | fp16 q4 q8 | 1.5 GB |
Falcon-3-TII |
2024-12-31 | Details |
|
|
Falcon3 10B Instruct | "Falcon3 10B: TII's 10 billion param, multilingual model for extensive text generation." | 10B | fp16 q4 q8 | 5.0 GB |
Falcon-3-TII |
2024-12-31 | Details |
| No Logo | Sailor2 8B | 9B | fp16 q4 q8 | 4.5 GB |
Apache-2.0 |
2024-12-31 | Details | |
| No Logo | Sailor2 20B | 19B | fp16 q4 q8 | 9.5 GB |
Apache-2.0 |
2024-12-31 | Details | |
|
|
Nous Hermes2 Mixtral 8X7B | Nous Hermes2 Mixtral 8x7B: 7B param, 2k context LLM by Nousresearch. Ideal for coding data visualizations. | 47B | fp16 q2 q3 q4 q5 q6 q8 | 23.5 GB |
Apache-2.0 Apache-2.0 |
2024-12-31 | Details |
| No Logo | Exaone3.5 7.8B Instruct | "Exaone3.5 7.8B: LG AI's 7.8 billion param, 32k context-length LLM for bilingual conversation & chatbot inference." | 8B | fp16 q4 q8 | 4.0 GB |
EXAONE-AIMLA-11-NC |
2024-12-31 | Details |
| No Logo | Exaone3.5 2.4B Instruct | "Exaone3.5 2.4B: LG AI's 2.4 billion param, 32k context-length LLM for bilingual conversation & chatbot inference." | 3B | fp16 q4 q8 | 1.5 GB |
EXAONE-AIMLA-11-NC |
2024-12-31 | Details |
| No Logo | Smollm2 135M Instruct | "Smollm2 135M Instruct by Hugging Face Smol Models Research Enterprise. A compact, 135M parameter model for English text rewriting & summarization." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-11-29 | Details |
| No Logo | Smollm2 360M Instruct | "Smollm2 360M Instruct by Hugging Face Smol Models Research Enterprise. A 360M parameter, mono-lingual model designed for text rewriting & summarization." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-11-29 | Details |
|
|
Mistral Large 123B Instruct | "Mistral Large 123B: A powerful, open-source, multilingual model by Mistral AI. With 123 billion parameters and a context window of 128k, it's ideal for research and non-commercial use." | 123B | fp16 q2 q3 q4 q5 q6 q8 | 61.5 GB |
Not Found Not Found |
2024-11-29 | Details |
| No Logo | Minicpm V 8B | "Minicpm V 8B: Open-source, 8 billion param. LLM for multi-lingual image classification tasks." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 Apache-2.0 Unknown |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 7B Base | "Qwen2.5 Coder 7B Base by Alibaba Qwen: A 7 billion parameter monolithic model for coding tasks." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 1.5B Base | "Qwen2.5 Coder 1.5B: Alibaba's 1.5B param model for coding, supports 32k context, excels in code generation." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 1.5B Instruct | "Qwen2.5 Coder 1.5B Instruct by Alibaba, a 1.5B parameter mono-lingual model with 32k context-length, excels in code generation." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 7B Instruct | "Qwen2.5 Coder 7B Instruct by Alibaba Qwen: A 7 billion parameter, mono-lingual large language model designed for efficient code generation with context lengths up to 128k." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 0.5B Base | "Qwen2.5 Coder 0.5B: Alibaba's 0.5B param, mono-lingual LLM for efficient code generation." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 3B Base | "Qwen2.5 Coder 3B: Alibaba's 3 billion param, mono-lingual LLM for efficient code generation." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Qwen-RESEARCH |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 14B Base | "Qwen2.5 Coder 14B Base by Alibaba Qwen: A versatile, multilingual large language model with 14 billion parameters, designed for efficient code generation and debugging." | 15B | fp16 q2 q3 q4 q5 q6 q8 | 7.5 GB |
Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 32B Base | "Qwen2.5 Coder 32B Base by Alibaba Qwen: A powerful 32 billion parameter large language model designed for monolingual code generation and improvement." | 33B | fp16 q2 q3 q4 q5 q6 q8 | 16.5 GB |
Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 0.5B Instruct | "Qwen2.5 Coder 0.5B Instruct by Alibaba Qwen: Multilingual LLM with 0.5B params, 32k/8k context-length, excels in generating lengthy texts." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 14B Instruct | "Qwen2.5 Coder 14B Instruct by Alibaba Qwen: 14 billion param, 32k/128k context, mono-lingual, excels in code generation." | 15B | fp16 q2 q3 q4 q5 q6 q8 | 7.5 GB |
Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 32B Instruct | "Qwen2.5 Coder 32B Instruct by Alibaba Qwen: 32 billion param, bilingual, excels in coding tasks." | 33B | fp16 q2 q3 q4 q5 q6 q8 | 16.5 GB |
Apache-2.0 |
2024-11-29 | Details |
| No Logo | Qwen2.5 Coder 3B Instruct | "Qwen2.5 Coder 3B Instruct: Alibaba's multilingual LLM with 3B params & 32k/8k context lengths, excelling in long text generation." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Qwen-RESEARCH |
2024-11-29 | Details |
|
|
Granite3 Dense 2B Instruct | "Granite3 Dense 2B: IBM's multilingual, 4k context-length model for efficient text summarization." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Apache-2.0 |
2024-11-29 | Details |
|
|
Granite3 Dense 8B Instruct | "Granite3 Dense 8B: IBM's 8 billion param, multilingual LLM. Ideal for text summarization with a 4k context window." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 |
2024-11-29 | Details |
|
|
Granite3 Moe 3B Instruct | "Granite3 Moe 3B: IBM's 3 billion param, multilingual LLM. Ideal for text summarization with a 4k context window." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Apache-2.0 |
2024-11-29 | Details |
|
|
Granite3 Moe 1B Instruct | "Granite3 Moe 1B: IBM's multilingual, 1 billion parameter model for efficient text summarization with a 4k context window." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
Apache-2.0 |
2024-11-29 | Details |
| No Logo | Smollm2 1.7B Instruct | "Smollm2 2B Instruct by Hugging Face TB Research Enterprise: A 2 billion parameter, mono-lingual LLM with a 2k context window, excelling in text rewriting tasks." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 |
2024-11-29 | Details |
|
|
Granite3 Guardian 2B | "Granite3 Guardian 3B: IBM's 3B param, 2k context-length LLM for monolingual harm detection in prompts and responses." | 3B | fp16 q5 q6 q8 | 1.5 GB |
Apache-2.0 |
2024-11-29 | Details |
|
|
Granite3 Guardian 8B | "Granite3 Guardian 8B: IBM's 8 billion param. LLM for harm risk detection in prompts/responses, with a 2k context window." | 8B | fp16 q5 q6 q8 | 4.0 GB |
Apache-2.0 |
2024-11-29 | Details |
|
|
Athene V2 72B | "Athene V2 73B by Nexusflow: A powerful, 73-billion parameter large language model designed for chat applications. Supports English with a context window of 2048 tokens." | 73B | fp16 q2 q3 q4 q5 q6 q8 | 36.5 GB |
Nexusflowai-Personal |
2024-11-29 | Details |
| No Logo | Opencoder 8B Instruct | "Opencoder 8B by Infly-Ai: 8 billion param. LLM for bilingual coding tasks, with context lengths up to 8k." | 8B | fp16 q4 q8 | 4.0 GB |
None |
2024-11-29 | Details |
| No Logo | Opencoder 1.5B Instruct | "Opencoder 1.5B by Infly-Ai: 1.5B params, 4k/8k context, bilingual support for efficient code generation." | 2B | fp16 q4 q8 | 1.0 GB |
None |
2024-11-29 | Details |
| No Logo | Llama3.2 Vision 11B Instruct | "Llama3.2 Vision 11B by Meta Llama Enterprise: 11 billion params, 128k context-length, supports multilingual languages. Expert in visual question answering & visual reasoning." | 10B | fp16 q4 q8 | 5.0 GB |
LLAMA-32-COMMUNITY |
2024-11-29 | Details |
| No Logo | Llama3.2 Vision 90B Instruct | "Llama3.2 Vision 90B: Meta's 90B param, multi-lingual LLM for visual QA & reasoning." | 88B | fp16 q4 q8 | 44.0 GB |
LLAMA-32-COMMUNITY |
2024-11-29 | Details |
|
|
Aya Expanse 8B | "Aya Expanse 8B by Cohere AI: An 8 billion parameter, multi-lingual LLM with an 8k context window, excelling as a versatile writing assistant." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
CC-BY-NC-4.0 |
2024-10-29 | Details |
|
|
Aya Expanse 32B | "Aya Expanse 32B by Cohere AI: A powerful, 32-billion parameter, multi-lingual language model with a 128k context window, designed as an advanced multilingual writing assistant." | 32B | fp16 q2 q3 q4 q5 q6 q8 | 16.0 GB |
CC-BY-NC-4.0 |
2024-10-29 | Details |
|
|
Nemotron 70B Instruct | "Nemotron 70B by Nvidia: 70 billion param. LLM for general-domain tasks, supports context lengths up to 128k." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
Llama-31-AUP LLAMA-31-CCLA |
2024-10-29 | Details |
|
|
Shieldgemma 9B | "Shieldgemma 9B: Google's 9 billion param. LLM, 4k context-length, English-only. Safely moderates user & model outputs." | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
Gemma-Terms-of-Use |
2024-10-29 | Details |
|
|
Shieldgemma 2B | "Shieldgemma 2B: Google's 2 billion param, 4k context-length LLM. Mono-lingual, designed for safe content moderation." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Gemma-Terms-of-Use |
2024-10-29 | Details |
|
|
Shieldgemma 27B | "Shieldgemma 27B: Google's 27 billion parameter LLM, designed for safe content moderation. Supports English with a context window of 4k." | 27B | fp16 q2 q3 q4 q5 q6 q8 | 13.5 GB |
Gemma-Terms-of-Use |
2024-10-29 | Details |
| No Logo | Llama Guard3 8B | "Llama Guard3 8B: Meta's 8 billion param, multi-lingual model for real-time LLM input/output safety classification." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
LLAMA-31-CCLA |
2024-10-29 | Details |
| No Logo | Llama Guard3 1B | "Llama Guard3 1B: Meta's 1 billion param, multi-lingual model for efficient content safety classification." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Llama-32-AUP LLAMA-32-COMMUNITY |
2024-10-29 | Details |
|
|
Qwen2 7B | Qwen2 8B: An 8 billion parameter, monolingual LLM by Qwen, excelling in natural language understanding with a context window of 2048 tokens. | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 TQ-LA Unlicense Apache-2.0 |
2024-09-29 | Details |
|
|
Qwen2 72B | Qwen2 72B: Qwen's 72 billion param, 2k context-length LLM for advanced language understanding. | 73B | fp16 q2 q3 q4 q5 q6 q8 | 36.5 GB |
Apache-2.0 TQ-LA Unlicense TQ-LA |
2024-09-29 | Details |
|
|
Qwen2 72B Instruct | "Qwen's Qwen2 72B Instruct: A powerful bi-lingual LLM with 72 billion parameters, designed for advanced language understanding and generation tasks." | 73B | fp16 q2 q3 q4 q5 q6 q8 | 36.5 GB |
Apache-2.0 TQ-LA Unlicense TQ-LA |
2024-09-29 | Details |
|
|
Qwen2 7B Instruct | "Qwen's Qwen2 7B Instruct: A 7 billion parameter, mono-lingual LLM for advanced language understanding." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 TQ-LA Unlicense Apache-2.0 |
2024-09-29 | Details |
|
|
Qwen2 1.5B Instruct | Qwen2 1.5B Instruct: Bi-lingual Large Language Model by Qwen, 5B params, excels in language understanding. | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 TQ-LA Unlicense Apache-2.0 |
2024-09-29 | Details |
|
|
Qwen2 0.5B Instruct | Qwen's Qwen2 0.5B Instruct: A 5 billion parameter, monolingual model for advanced language understanding. | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 TQ-LA Unlicense Apache-2.0 |
2024-09-29 | Details |
| No Logo | Llama3.1 8B | "Llama3.1 8B: Meta's 8 billion param, 128k context-length LLM for multilingual assistant chat." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
LLAMA-31-CCLA LLAMA-31-CCLA |
2024-09-29 | Details |
|
|
Deepseek Coder V2 16B Base | "Deepseek Coder V2 16B: 16 billion param, 128k context-length model by Deepseek for monolingual code completion." | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
DEEPSEEK-LICENSE MIT DEEPSEEK-LICENSE MIT DEEPSEEK-LICENSE |
2024-09-29 | Details |
| No Logo | Llama3.1 405B | "Llama3.1 405B by Meta Llama: A powerful, multilingual large language model with a context window of 128k, designed for assistant-like chat apps." | 406B | fp16 q2 q3 q4 q5 q6 q8 | 203.0 GB |
LLAMA-31-CCLA LLAMA-31-CCLA |
2024-09-29 | Details |
| No Logo | Llama3.1 405B Instruct | "Llama3.1 405B Instruct by Meta Llama: Multilingual Assistant Chat, 405B Parameters, 128K Context Length." | 406B | fp16 q2 q3 q4 q5 q6 q8 | 203.0 GB |
LLAMA-31-CCLA LLAMA-31-CCLA |
2024-09-29 | Details |
| No Logo | Llama3.1 8B Instruct | "Llama3.1 8B Instruct by Meta Llama, an 8 billion parameter, multi-lingual model with a 128k context window, excels in assistant-like chat." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
LLAMA-31-CCLA LLAMA-31-CCLA |
2024-09-29 | Details |
|
|
Yi Coder 1.5B Base | "Yi Coder 1.5B Base by 01-AI: A 1.5 billion parameter monolithic model for coding assistance with variable context lengths up to 8k." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
Apache-2.0 |
2024-09-29 | Details |
|
|
Yi Coder 9B Base | "Yi Coder 9B Base by 01-AI: A 9 billion parameter LLM for coding assistance with variable context lengths up to 8k." | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
Apache-2.0 |
2024-09-29 | Details |
|
|
Deepseek Coder V2 236B Base | "Deepseek Coder V2: 236B params, 128k context. Mono-lingual, excels in code completion." | 236B | fp16 q2 q3 q4 q5 q6 q8 | 118.0 GB |
DEEPSEEK-LICENSE MIT DEEPSEEK-LICENSE MIT DEEPSEEK-LICENSE |
2024-09-29 | Details |
|
|
Deepseek Coder V2 236B Instruct | "Deepseek Coder V2: 236B Instruct, a large language model by Deepseek for code completion, supports English with a context-length of 128k." | 236B | fp16 q2 q3 q4 q5 q6 q8 | 118.0 GB |
DEEPSEEK-LICENSE MIT MIT DEEPSEEK-LICENSE |
2024-09-29 | Details |
|
|
Deepseek Coder V2 16B Instruct | "Deepseek Coder V2 16B Instruct: 16B params, 128k context. Mono-lingual, excels in code completion." | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
DEEPSEEK-LICENSE MIT MIT DEEPSEEK-LICENSE |
2024-09-29 | Details |
| No Logo | Llama3.1 70B | "Llama3.1 70B by Meta Llama: 70 billion params, 128k context-length, multilingual support for assistant-style chats." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
LLAMA-31-CCLA LLAMA-31-CCLA |
2024-09-29 | Details |
| No Logo | Llama3.1 70B Instruct | "Llama3.1 70B Instruct by Meta Llama: Multilingual Assistant Chat, 70B params, 128k context." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
LLAMA-31-CCLA LLAMA-31-CCLA |
2024-09-29 | Details |
|
|
Phi3.5 3.8B Instruct | "Phi3.5 4B Instruct by Microsoft: A compact, 4 billion parameter model with 4k to 128k context-length, ideal for resource-constrained AI systems." | 4B | fp16 q2 q3 q4 q5 q6 q8 | 2.0 GB |
MIT MIT |
2024-09-29 | Details |
|
|
Codestral 22B | Codestral 22B Instruct by Mistral AI: 22B param model for coding tasks. Supports English, answers queries, generates docs, explains/factorizes code. | 22B | q2 q3 q4 q5 q6 q8 | 11.0 GB |
Mistral-AI-NPL Mistral-AI-NPL |
2024-09-29 | Details |
|
|
Reflection 70B | None | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
LLAMA-31-CCLA LLAMA-31-CCLA |
2024-09-29 | Details |
|
|
Deepseek V2.5 236B | "Deepseek V2.5 236B: Bi-lingual, 8k context-length LLM for coding in multiple languages." | 236B | q4 q5 q8 | 118.0 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2024-09-29 | Details |
|
|
Yi Coder 9B | "Yi Coder 9B by 01-AI: A powerful, open-source, monolingual large language model with 9 billion parameters and a context length of 128k. Primarily designed to assist in coding tasks." | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
|
|
Yi Coder 1.5B | "Yi Coder 1.5B by 01-AI: 1.5B params, 128k context, mono-lingual, excels in coding assistance." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
| No Logo | Reader Lm 1.5B | "ReaderLM 1.5B by Jina AI: A 1.5 billion parameter, monolingual language model with a 256k context window, designed for content conversion tasks." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
CC-BY-NC-4.0 CC-BY-NC-4.0 |
2024-09-29 | Details |
| No Logo | Reader Lm 0.5B | "ReaderLM 0.5B by Jina AI: A 0.5 billion parameter, monolingual language model with a 256k context window, designed for content conversion tasks." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
CC-BY-NC-4.0 CC-BY-NC-4.0 |
2024-09-29 | Details |
| No Logo | Qwen2.5 0.5B Instruct | "Qwen2.5 0.5B Instruct by Alibaba Qwen: Multilingual LLM with 32k/8k context-length for extended text generation." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
| No Logo | Qwen2.5 1.5B Instruct | "Qwen2.5 Instruct: Alibaba's 1.5B param model for multilingual long text generation up to 32k tokens." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
| No Logo | Qwen2.5 3B Instruct | "Qwen2.5 3B Instruct: Alibaba's multilingual LLM with 3B params, 32k/8k context-length, excels in long text generation." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Qwen-RESEARCH Qwen-RESEARCH |
2024-09-29 | Details |
| No Logo | Qwen2.5 7B Instruct | "Qwen2.5 7B Instruct: Alibaba's multilingual LLM with 7 billion params, supports context lengths up to 128k for long text generation." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
| No Logo | Qwen2.5 32B Instruct | "Qwen2.5 32B Instruct: Alibaba's multilingual LLM with 32 billion params, supporting context lengths up to 128k for long text generation." | 33B | fp16 q2 q3 q4 q5 q6 q8 | 16.5 GB |
Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
| No Logo | Nemotron Mini 4B Instruct | "NVIDIA's Nemotron Mini 4B: A 4 billion parameter, monolingual LLM with a 4k context window, designed by NVIDIA Enterprise for creative roleplaying response generation." | 4B | fp16 q2 q3 q4 q5 q6 q8 | 2.0 GB |
NAIFMCLA NAIFMCLA |
2024-09-29 | Details |
|
|
Solar Pro 22B Instruct | "Solar Pro 22B by Upstage: A 22 billion parameter, English-only LLM with a 4096 token context window, expertly tuned for instruction-following and conversation." | 22B | fp16 q2 q3 q4 q5 q6 q8 | 11.0 GB |
MIT MIT |
2024-09-29 | Details |
| No Logo | Bespoke Minicheck 7B | "Bespoke Minicheck 7B: A 7 billion parameter, mono-lingual LLM by Bespoke Labs. Ideal for fact-checking with a 32k context window." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB | 2024-09-29 | Details | |
| No Logo | Llama3.2 3B | "Llama3.2 3B: Meta's 3 billion param, multi-lingual LLM for assistant chat apps." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Llama-32-AUP LLAMA-32-COMMUNITY Llama-32-AUP LLAMA-32-COMMUNITY |
2024-09-29 | Details |
| No Logo | Llama3.2 1B | "Llama3.2 1B: Meta's 1B param model for multilingual chat apps, with context lengths up to 128k." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
Llama-32-AUP LLAMA-32-COMMUNITY Llama-32-AUP LLAMA-32-COMMUNITY |
2024-09-29 | Details |
| No Logo | Llama3.2 1B Instruct | "Llama3.2 1B Instruct by Meta Llama Enterprise: A versatile, multilingual large language model with 1 billion parameters and context lengths up to 128k, designed for assistant-like chat applications." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
Llama-32-AUP LLAMA-32-COMMUNITY Llama-32-AUP LLAMA-32-COMMUNITY |
2024-09-29 | Details |
| No Logo | Llama3.2 3B Instruct | "Llama3.2 3B Instruct by Meta Llama Enterprise: A multilingual, 3 billion parameter model with 8k to 128k context-length, designed for assistant-like chat apps." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Llama-32-AUP LLAMA-32-COMMUNITY Llama-32-AUP LLAMA-32-COMMUNITY |
2024-09-29 | Details |
| No Logo | Qwen2.5 14B Instruct | "Qwen2.5 14B Instruct: Alibaba's multilingual LLM with 14 billion params, supporting context lengths up to 128k for extensive text generation." | 15B | fp16 q2 q3 q4 q5 q6 q8 | 7.5 GB |
Apache-2.0 |
2024-09-29 | Details |
| No Logo | Qwen2.5 72B Instruct | "Qwen2.5 72B Instruct: Alibaba's 72 billion parameter LLM, supports multilingual text generation up to 128k tokens." | 73B | fp16 q2 q3 q4 q5 q6 q8 | 36.5 GB |
Qwen-RESEARCH |
2024-09-29 | Details |
|
|
Starcoder2 3B | "Starcoder2 3B by Bigcodeproject: 3 billion params, 16k/4k context, mono-lingual, excels in code gen & completion." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
BigCode-Open-RAIL-M-v1 BigCode-Open-RAIL-M-v1 |
2024-09-29 | Details |
|
|
Starcoder2 15B | "Starcoder2 15B by Bigcodeproject: 15 billion param. LLM for coding tasks, supports 16k/4k context lengths, generates code snippets in multiple languages." | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
BigCode-Open-RAIL-M-v1 BigCode-Open-RAIL-M-v1 |
2024-09-29 | Details |
|
|
Starcoder2 7B | "Starcoder2 7B by Bigcodeproject: 7 billion params, 16k/4k context, mono-lingual. Expert in generating code snippets." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
BigCode-Open-RAIL-M-v1 BigCode-Open-RAIL-M-v1 |
2024-09-29 | Details |
|
|
Starcoder2 15B Instruct | "Starcoder2 15B Instruct by Bigcodeproject: A large language model with 15 billion parameters, designed for Python code generation. It offers a context-length of up to 4k and is monolingual." | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
BigCode-Open-RAIL-M-v1 BigCode-Open-RAIL-M-v1 |
2024-09-29 | Details |
|
|
Granite Code 3B Base | "Granite Code 3B: IBM's 3 billion param, 2k context-length model for monolingual code generation." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Apache-2.0 Unknown Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
|
|
Granite Code 34B Base | Granite Code 34B: IBM's 34 billion param, 8k context-length model for monolingual code generation. | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
Apache-2.0 Unknown Apache-2.0 |
2024-09-29 | Details |
|
|
Granite Code 20B Base | Granite Code 20B: IBM's 20 billion param, monolingual model for efficient code generation. | 20B | fp16 q2 q3 q4 q5 q6 q8 | 10.0 GB |
Apache-2.0 Unknown Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
|
|
Granite Code 8B Base | Granite Code 8B: IBM's 8 billion param, 4k context-length model for monolingual code generation. | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 Unknown Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
| No Logo | Qwen2.5 0.5B Base | "Qwen2.5 Base: Alibaba's 0.5B param LLM for multilingual long text generation." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-09-29 | Details |
|
|
Granite Code 34B Instruct | "Granite Code 34B Instruct: IBM's 34B param, mono-lingual LLM. Ideal for coding assistance." | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
Apache-2.0 Unknown Apache-2.0 |
2024-09-29 | Details |
|
|
Granite Code 20B Instruct | "Granite Code 20B Instruct by IBM: A 20 billion parameter, monolingual LLM with an 8k context window, designed for crafting advanced coding assistants." | 20B | fp16 q2 q3 q4 q5 q6 q8 | 10.0 GB |
Apache-2.0 Unknown Apache-2.0 |
2024-09-29 | Details |
|
|
Granite Code 8B Instruct | "Granite Code 8B Instruct: IBM's 8 billion param, 4k context-length model for monolingual coding assistance." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 Unknown Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
|
|
Granite Code 3B Instruct | "Granite Code 3B Instruct by IBM: A powerful, open-source large language model with 3 billion parameters and a context window of 2048 tokens. Designed for building coding assistants in English." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Apache-2.0 Unknown Apache-2.0 Apache-2.0 |
2024-09-29 | Details |
| No Logo | Bge M3 567M | "Bge M3 567M by BAAI: A bi-lingual, 567M param model for efficient information retrieval with an 8k context window." | 1B | fp16 | 0.5 GB |
MIT |
2024-08-29 | Details |
| No Logo | Smollm 1.7B Base | "Smollm 1.7B Base by Hugging Face Smol Models Research Enterprise: A 1.7 billion parameter, monolingual language model with a context-length of 4k, designed as an assistive tool for versatile text generation." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 |
2024-08-29 | Details |
| No Logo | Smollm 135M Base | "Smollm 135M Base by Hugging Face Smol Models Research Enterprise. A compact, 135M parameter model with a context-length of 4k, designed as an assistive tool for multilingual text generation across diverse topics." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-08-29 | Details |
|
|
Mistral Nemo 12B Instruct | "Mistral Nemo 12B: NVIDIA's multilingual, 12 billion parameter model for advanced text completion and generation tasks." | 12B | fp16 q2 q3 q4 q5 q6 q8 | 6.0 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
| No Logo | Smollm 135M Instruct | "Smollm 135M Instruct by Hugging Face, a compact monolingual model for answering general knowledge queries." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-08-29 | Details |
| No Logo | Smollm 360M Base | "Smollm 360M Base by Hugging Face Smol Models Research Enterprise. A compact, 360M parameter model with 4k context-length, designed as an efficient assistive tool for multilingual text generation." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-08-29 | Details |
| No Logo | Smollm 360M Instruct | "Smollm 360M Instruct by Hugging Face, a 360M param model for general knowledge queries." | 0B | fp16 q2 q3 q4 q5 q6 q8 | 0.0 GB |
Apache-2.0 |
2024-08-29 | Details |
| No Logo | Internlm2 7B | "InternLM's Internlm2 8B: An 8 billion parameter, bi-lingual large language model with a 200k context window, designed for downstream deep adaptations." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
| No Logo | Bge Large 335M | "Beijing AI's Bge Large 0B: Bi-lingual model for efficient text retrieval." | 0B | fp16 | 0.0 GB |
MIT MIT |
2024-08-29 | Details |
| No Logo | Paraphrase Multilingual 278M Base | "Paraphrase Multilingual 0B by Sentence Transformers: Bi-lingual, 0B params, 0k context. Ideal for sentence/paragraph clustering." | 0B | fp16 | 0.0 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
| No Logo | Smollm 1.7B Instruct | "Smollm 1.7B Instruct by Hugging Face TB Research, a 7B parameter model with 2k context-length, excels in answering general knowledge questions in English." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
| No Logo | Internlm2 1.8B | "InternLM's Internlm2, a 1.8B parameter model with 200k context-length, excels in downstream tasks, supporting monolingual applications." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
| No Logo | Internlm2 20B | "InternLM's Internlm2 20B: A 20 billion parameter, bi-lingual large language model with a 200k context-length, designed for deep domain adaptation." | 20B | fp16 q2 q3 q4 q5 q6 q8 | 10.0 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
| No Logo | Qwen2 Math 7B Instruct | "Qwen2 Math 7B: Alibaba's 7 billion parameter, monolingual LLM for efficient math problem-solving." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
| No Logo | Qwen2 Math 1.5B Instruct | "Qwen2 Math 1.5B: Alibaba's 5B param model for math problem-solving, supports English, context window of 2k." | 2B | fp16 q2 q3 q4 q5 q6 q8 | 1.0 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
| No Logo | Qwen2 Math 72B Instruct | "Qwen2 Math 72B: Alibaba's 72 billion param, monolingual LLM for math problem-solving." | 73B | fp16 q2 q3 q4 q5 q6 q8 | 36.5 GB |
Apache-2.0 Apache-2.0 |
2024-08-29 | Details |
|
|
Command R Plus 104B | "Command R Plus 104B by Cohere AI: A 104 billion parameter, multi-lingual large language model designed for complex reasoning tasks with a context-length of 128k." | 104B | fp16 q2 q3 q4 q5 q6 q8 | 52.0 GB |
CC-BY-NC-4.0 CC-BY-NC-4.0 CC-BY-NC-4.0 |
2024-08-29 | Details |
|
|
Command R 35B | "Command R 35B by Cohere AI: A powerful, multilingual large language model with a 128k context window, designed for advanced reasoning tasks." | 35B | fp16 q2 q3 q4 q5 q6 q8 | 17.5 GB |
CC-BY-NC-4.0 |
2024-08-29 | Details |
|
|
Codellama 70B | "Codellama 70B: A 70 billion parameter large language model by Code Llama, designed for monolingual code completion tasks with a context length of up to 10,000 tokens." | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
LLAMA-2-CLA |
2024-07-29 | Details |
|
|
Gemma2 9B | Gemma2 9B: Google's 9 billion parameter LLM for creative content generation. Supports English with context lengths up to 32K. | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
| No Logo | Glm4 9B | "Glm4 9B: A versatile, multi-lingual large language model by Tsinghua University's KEG & Data Mining group. With 9 billion parameters and dynamic context lengths up to 10,240 tokens, it excels in content generation tasks." | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
GLM-4-9B |
2024-07-29 | Details |
|
|
Gemma2 27B | Gemma2 27B: Google's 27 billion parameter LLM for creative content generation. Supports English with context lengths up to 32K. | 27B | fp16 q2 q3 q4 q5 q6 q8 | 13.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
|
|
Gemma2 2B | "Gemma2 2B: Google's 2 billion parameter, monolingual LLM for creative content & communication. Generates texts up to 8k tokens." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
|
|
Gemma2 27B Instruct | Gemma2 27B Instruct: Google's 27 billion parameter LLM for creative content generation. Supports English with context lengths up to 32K. | 27B | fp16 q2 q3 q4 q5 q6 q8 | 13.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
|
|
Gemma2 9B Instruct | Gemma2 9B Instruct: Google's 9 billion parameter, 2k context-length model for monolingual content creation & communication. | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
|
|
Gemma2 2B Instruct | "Gemma2 2B Instruct by Google: A powerful 2 billion parameter LLM for creative content generation & communication. Supports English with an 8k context window." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
| No Logo | Glm4 9B | "Glm4 9B: A 9 billion parameter, 32k context-length model by Tsinghua University's KEG & Data Mining group. Ideal for creating long-form content like detailed travel guides and comprehensive reports." | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
GLM-4-9B GLM-4-9B |
2024-07-29 | Details |
| No Logo | Codegeex4 9B | "Codegeex4 9B: Bi-lingual, 128k context-length model by Tsinghua University for code completion & generation." | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
GLM-4-9B GLM-4-9B |
2024-07-29 | Details |
| No Logo | Llama3 Groq Tool Use 8B | Llama3 Groq Tool Use 8B: An 8 billion parameter, monolingual large language model by Groq Inc., designed for advanced tool use tasks with a context window of 2048 tokens. | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
META-LLAMA-3-GROQ META-LLAMA-3-GROQ |
2024-07-29 | Details |
| No Logo | Llama3 Groq Tool Use 70B | Llama3 Groq Tool Use 70B: 70 billion param, 2k context-length model by Groq Inc. Specializes in tool use & function calling tasks. | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
META-LLAMA-3-GROQ META-LLAMA-3-GROQ |
2024-07-29 | Details |
|
|
Mathstral 7B | Mathstral 7B Text is an advanced open-source language model developed by Mistral AI, a leading company in the field. With 7 billion parameters, it offers sophisticated and nuanced understanding of text, making it a powerful tool for natural language processing tasks. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 Apache-2.0 |
2024-07-29 | Details |
| No Logo | Firefunction V2 70B | "Firefunction V2 70B by Fireworks AI: A powerful, 70 billion parameter model with an 8k context window, designed for general instruction following in English." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-07-29 | Details |
| No Logo | Nuextract 3.8B | "Nuextract 4B: NuMind's 4 billion parameter, 2k context-length LLM for efficient monolingual info extraction." | 4B | fp16 q2 q3 q4 q5 q6 q8 | 2.0 GB |
Apache-2.0 Apache-2.0 |
2024-07-29 | Details |
|
|
Phi3 3.8B Instruct | "Phi3 4B Instruct by Microsoft: A compact (4B params), efficient (4k-128k context) monolingual model for resource-constrained AI systems." | 4B | fp16 q2 q3 q4 q5 q6 q8 | 2.0 GB |
Microsoft Microsoft |
2024-07-29 | Details |
|
|
Phi3 14B Instruct | "Phi3 14B Instruct by Microsoft: A compact, high-performance LLM with 14 billion parameters, supporting context lengths up to 128k. Ideal for resource-constrained AI systems." | 14B | fp16 q2 q3 q4 q5 q6 q8 | 7.0 GB |
Microsoft Microsoft |
2024-07-29 | Details |
|
|
Mistral 7B | Mistral 7B: A 7 billion parameter, monolingual language model by Mistral AI, excelling in text generation with a context window of 2048 tokens. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 Apache-2.0 |
2024-07-29 | Details |
|
|
Mistral 7B Instruct | "Mistral 7B Instruct: A 7 billion parameter LLM by Mistral AI, supporting up to 32k context length. Ideal for creating human-like text across diverse applications." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 Apache-2.0 |
2024-07-29 | Details |
|
|
Codegemma 7B | Codegemma 7B by Google: 7 billion params, 2k context, mono-lingual, excels in code completion. | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
|
|
Codegemma 2B | CodeGemma 2B Text is an open-source large language model developed by Google, designed to understand and generate human-like text with a parameter size of 2 billion. | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
|
|
Codegemma 7B Instruct | Codegemma 7B Instruct by Google: 7 billion params, 2k context, mono-lingual, excels in code completion. | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-07-29 | Details |
|
|
Codellama 7B | Codellama 7B: A 7 billion parameter, mono-lingual large language model by Code Llama, excelling in code completion with a context-length of 4096 tokens. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-CLA Llama-CODE-AUP Llama-CODE-AUP LLAMA-2-CLA |
2024-07-29 | Details |
|
|
Codellama 34B | Codellama 34B: 34 billion param. LLM by Code Llama, excelling in code completion with a 4k context window. | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
LLAMA-2-CLA Llama-CODE-AUP Llama-CODE-AUP LLAMA-2-CLA |
2024-07-29 | Details |
|
|
Codellama 13B | Codellama 13B: A 13 billion parameter, monolingual large language model by Code Llama, designed for efficient code completion with a context-length of 4096 tokens. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
LLAMA-2-CLA Llama-CODE-AUP Llama-CODE-AUP LLAMA-2-CLA |
2024-07-29 | Details |
|
|
Codellama 7B Instruct | Codellama 7B Instruct: A 7 billion parameter, mono-lingual large language model by Code Llama, designed for efficient code completion tasks with a context window of 2048 tokens. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-CLA Llama-CODE-AUP Llama-CODE-AUP LLAMA-2-CLA |
2024-07-29 | Details |
|
|
Codellama 70B Instruct | "Codellama 70B Instruct: A 70 billion parameter, mono-lingual LLM by Code Llama, excelling in code synthesis & understanding tasks with a 16k context window." | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
LLAMA-2-CLA Llama-CODE-AUP |
2024-07-29 | Details |
|
|
Codellama 34B Instruct | Codellama 34B Instruct: A 34 billion parameter, mono-lingual large language model by Code Llama, designed for advanced code completion tasks with a context-length of 2048 tokens. | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
LLAMA-2-CLA Llama-CODE-AUP |
2024-07-29 | Details |
|
|
Codellama 13B Instruct | Codellama 13B Instruct by Code Llama. 13 billion params, 2k context-length. Mono-lingual, excels in code completion. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
LLAMA-2-CLA Llama-CODE-AUP Llama-CODE-AUP LLAMA-2-CLA |
2024-07-29 | Details |
|
|
Deepseek V2 16B | "Deepseek V2 16B: Bi-lingual, 16 billion param, 128k context-length LLM by Deepseek, excelling in code generation." | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2024-06-29 | Details |
|
|
Deepseek V2 236B | "Deepseek V2 236B: Bi-lingual, 236 billion param. LLM for efficient code generation." | 236B | fp16 q2 q3 q4 q5 q6 q8 | 118.0 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2024-06-29 | Details |
|
|
Codeqwen 7B | "CodeQwen 7B: A 7 billion parameter, mono-lingual large language model by Qwen, excelling in code generation with a context-length of 64k." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
TQ-LA Unlicense TQ-LA |
2024-06-29 | Details |
|
|
Falcon2 11B | "Falcon2 11B: TII's multilingual, 11 billion parameter LLM with context lengths up to 8K. Ideal for large language model research." | 11B | fp16 q2 q3 q4 q5 q6 q8 | 5.5 GB |
Falcon-2-11B-TII-License-10 Falcon-2-11B-TII-License-10 |
2024-05-29 | Details |
| No Logo | Stablelm2 12B | "StableLM2 12B by Stability AI: A 12 billion parameter, monolingual language model for text generation with a context length of 4k." | 12B | q2 q3 q32 q4 q5 q6 q8 | 6.0 GB |
STABILITY-AI-NCRCLA |
2024-05-29 | Details |
| No Logo | Llama3 8B | "Llama3 8B: Meta's 8 billion param, 8k context-length model for English commercial & research use." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-05-29 | Details |
| No Logo | Llama3 70B | "Llama3 70B by Meta Llama: 70 billion params, 8k context-length, English-only. Ideal for commercial & research use." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-05-29 | Details |
| No Logo | Llama3 8B Instruct | "Llama3 8B Instruct by Meta Llama: An 8 billion parameter, English-focused model with an 8k context window, ideal for commercial applications." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-05-29 | Details |
| No Logo | Llama3 70B Instruct | "Llama3 70B Instruct by Meta Llama: A powerful, English-focused model with a 8k context window for commercial applications." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-05-29 | Details |
|
|
Aya 8B | "Aya 8B by Cohere AI: An 8 billion parameter, multi-lingual language model for diverse text generation." | 8B | q2 q3 q4 q5 q6 q8 | 4.0 GB |
CC-BY-NC-4.0 CC-BY-NC-4.0 |
2024-05-29 | Details |
|
|
Aya 35B | "Aya 35B by Cohere AI: 35 billion param, 8k context-length LLM for multilingual text generation." | 35B | q2 q3 q4 q5 q6 q8 | 17.5 GB |
CC-BY-NC-4.0 CC-BY-NC-4.0 |
2024-05-29 | Details |
|
|
Yi 6B | "Yi 6B: Bi-lingual LLM by 01-AI. Offers context lengths up to 200k for personalized use." | 6B | fp16 q2 q3 q4 q5 q6 q8 | 3.0 GB |
Apache-2.0 YSMCLA YSMLA Apache-2.0 YSMLA YSMCLA |
2024-05-29 | Details |
|
|
Yi 34B | "Yi 34B by 01-AI: 34 billion param. LLM with 4k, 32k, & 200k context lengths. Bi-lingual support for personal use." | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
Apache-2.0 YSMCLA YSMLA Apache-2.0 YSMCLA YSMLA |
2024-05-29 | Details |
|
|
Yi 9B | "Yi 9B: 01-AI's 9 billion parameter LLM. Supports English & Chinese. Offers context lengths up to 200k for versatile personal use." | 9B | fp16 q2 q3 q4 q5 q6 q8 | 4.5 GB |
YSMCLA YSMLA Apache-2.0 Apache-2.0 |
2024-05-29 | Details |
| No Logo | Stablelm2 1.6B | "StableLM2 1.6B by Stability AI: A 6 billion parameter, monolingual language model with a 4096 context window, designed as a versatile foundation for application-specific fine-tuning." | 2B | q2 q3 q32 q4 q5 q6 q8 | 1.0 GB |
STABILITY-AI-NCRCLA STABILITY-AI-NCRCLA STABILITY-AI-NCRCLA STABILITY-AI-NCRCLA STABILITY-AI-NCRCLA |
2024-05-29 | Details |
| No Logo | Stablelm2 12B | "StableLM2 12B by Stability AI: A 12-billion parameter, monolingual language model for text generation with a context length of up to 4096 tokens." | 12B | q2 q3 q32 q4 q5 q6 q8 | 6.0 GB |
STABILITY-AI-NCRCLA STABILITY-AI-NCRCLA |
2024-05-29 | Details |
| No Logo | Llama3 Gradient 8B Instruct | "Llama3 Gradient 8B by Meta Llama: An 8 billion parameter, English-focused LLM with an 8k context window, ideal for commercial & research use." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-05-29 | Details |
| No Logo | Llama3 Gradient 70B Instruct | "Llama3 Gradient 70B by Meta Llama: 70 billion params, 8k context-length, English-only. Ideal for commercial & research use." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-05-29 | Details |
|
|
Llama3 Chatqa 8B | Llama3 ChatQA 8B by NVIDIA: 8 billion param, 2k context-length model for monolingual conversational Q&A. | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-05-29 | Details |
|
|
Llama3 Chatqa 70B | Llama3 ChatQA 71B by NVIDIA: 71 billion param. model for conversational Q&A, supports English with context lengths up to 4K. | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-05-29 | Details |
|
|
Llava Llama3 8B | Llava Llama3 8B by Intel: 8 billion param., 2k context, mono-lingual, ideal for multimodal benchmarks. | 8B | fp16 q4 | 4.0 GB | 2024-05-29 | Details | |
| No Logo | Moondream 1.8B | Moondream 1B: Open-source, 1 billion param LLM. 4k context-length for image captioning in English. | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
Apache-2.0 Apache-2.0 |
2024-05-29 | Details |
|
|
Llava Phi3 3.8B | "Llava Phi3 4B: Xtuner's 4 billion param, monolingual model excels in image descriptions with a 4k context window." | 4B | fp16 q4 | 2.0 GB | 2024-05-29 | Details | |
| No Logo | Qwen 32B | "Qwen 32B: Alibaba's 32 billion parameter LLM for bilingual text generation, optimized with SFT, RLHF, and continued pretraining." | 32B | fp16 q2 q3 q32 q4 q5 q6 q8 | 16.0 GB |
TQ-LA |
2024-04-29 | Details |
| No Logo | Qwen 4B | "Qwen 4B: Alibaba's 4 billion parameter bilingual LLM for advanced text generation tasks." | 4B | fp16 q2 q3 q32 q4 q5 q6 q8 | 2.0 GB |
TQRLA |
2024-04-29 | Details |
| No Logo | Qwen 72B | "Qwen 72B: Alibaba's 72 billion parameter LLM for bilingual coding assistance." | 72B | fp16 q2 q3 q32 q4 q5 q6 q8 | 36.0 GB |
TQ-LA TQ-LA |
2024-04-29 | Details |
| No Logo | Qwen 7B | "Qwen 7B: Alibaba's 7 billion param, bi-lingual LLM for efficient code generation & debugging." | 8B | fp16 q2 q3 q32 q4 q5 q6 q8 | 4.0 GB |
TQ-LA |
2024-04-29 | Details |
|
|
Gemma 2B | "Gemma 2B: Google's 2 billion parameter LLM for creative text generation & chatbot AI." | 3B | fp16 q2 q3 q32 q4 q5 q6 q8 | 1.5 GB |
Gemma-Terms-of-Use |
2024-04-29 | Details |
|
|
Gemma 7B | "Gemma 7B: Google's 7 billion parameter LLM for creative text generation & chatbot development." | 9B | fp16 q2 q3 q32 q4 q5 q6 q8 | 4.5 GB |
Gemma-Terms-of-Use |
2024-04-29 | Details |
|
|
Snowflake Arctic Embed 110M | "Snowflake's Arctic Embed 110M: A 110M param, 8k context-length LLM by Snowflake, excelling in information retrieval." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-04-29 | Details |
|
|
Snowflake Arctic Embed 137M | "Snowflake's Arctic Embed 137M: A 137M param, monolingual LLM optimized for efficient information retrieval tasks." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-04-29 | Details |
|
|
Snowflake Arctic Embed 22M | "Snowflake's Arctic Embed 22M: A 22M parameter, monolingual model optimized for efficient information retrieval tasks." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-04-29 | Details |
|
|
Snowflake Arctic Embed 335M | "Snowflake Arctic Embed 335M: A 335M parameter, monolingual LLM by Snowflake. Ideal for info retrieval & SEO." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-04-29 | Details |
|
|
Snowflake Arctic Embed 33M | "Snowflake's Arctic Embed 33M: A 33M parameter, monolingual model optimized for efficient information retrieval tasks." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-04-29 | Details |
|
|
Dolphincoder 15B | Dolphincoder 15B: Community-maintained, 15 billion param. LLM for coding assistance, supports 32k context. | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
BigCode-Open-RAIL-M-v1 BigCode-Open-RAIL-M-v1 |
2024-04-29 | Details |
|
|
Gemma 7B | "Gemma 7B: Google's 7 billion parameter LLM for creative content & chat. Supports English, generates texts up to 8k tokens." | 9B | fp16 q2 q3 q32 q4 q5 q6 q8 | 4.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-04-29 | Details |
|
|
Gemma 2B | "Gemma 2B: Google's 2 billion param, 8k context-length model for monolingual English content creation & communication." | 3B | fp16 q2 q3 q32 q4 q5 q6 q8 | 1.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-04-29 | Details |
|
|
Gemma 7B Instruct | Gemma 7B Instruct by Google: A 7 billion parameter, monolingual LLM with an 8k context window, designed for creative content generation and effective communication. | 9B | fp16 q2 q3 q32 q4 q5 q6 q8 | 4.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-04-29 | Details |
|
|
Gemma 2B Instruct | Gemma 2B Instruct by Google: A powerful, 8k context-length model for monolingual content creation and communication. | 3B | fp16 q2 q3 q32 q4 q5 q6 q8 | 1.5 GB |
Gemma-Terms-of-Use Gemma-Terms-of-Use |
2024-04-29 | Details |
|
|
Qwen 4B | "Qwen 4B: Bi-lingual LLM with 4 billion params & 32k context-length. Ideal for text gen tasks post-training." | 4B | fp16 q2 q3 q32 q4 q5 q6 q8 | 2.0 GB |
TQRLA TQ-LA TQRLA TQRLA TQRLA |
2024-04-29 | Details |
|
|
Qwen 110B | "Qwen 110B: Bi-lingual LLM with 32k context-length. Ideal for advanced text generation tasks." | 111B | fp16 q2 q3 q32 q4 q5 q6 q8 | 55.5 GB |
TQRLA TQ-LA TQ-LA |
2024-04-29 | Details |
|
|
Qwen 72B | "Qwen 72B: Bi-lingual LLM with 72 billion params, 32k context-length for advanced NLP tasks." | 72B | fp16 q2 q3 q32 q4 q5 q6 q8 | 36.0 GB |
TQRLA TQ-LA TQ-LA TQ-LA TQ-LA TQ-LA |
2024-04-29 | Details |
|
|
Qwen 32B | "Qwen 32B: A 32-billion parameter, bilingual LLM with a 32K context window, optimized for advanced text generation tasks." | 32B | fp16 q2 q3 q32 q4 q5 q6 q8 | 16.0 GB |
TQRLA TQ-LA TQ-LA |
2024-04-29 | Details |
|
|
Qwen 14B | "Qwen 14B: Bi-lingual LLM with 2k/32k context, excelling in NLP tasks." | 14B | fp16 q2 q3 q32 q4 q5 q6 q8 | 7.0 GB |
TQRLA TQ-LA TQ-LA TQ-LA TQ-LA TQ-LA |
2024-04-29 | Details |
|
|
Qwen 7B | "Qwen 7B: Bi-lingual LLM with 7 billion params, supports context lengths up to 32k for diverse NLP tasks." | 8B | fp16 q2 q3 q32 q4 q5 q6 q8 | 4.0 GB |
TQRLA TQ-LA TQ-LA TQ-LA TQ-LA TQ-LA |
2024-04-29 | Details |
|
|
Qwen 1.8B | "Qwen 1.8B: Bi-lingual LLM with 8B params & 8k context, excelling in NLP tasks." | 2B | fp16 q2 q3 q32 q4 q5 q6 q8 | 1.0 GB |
TQRLA TQ-LA TQRLA TQRLA TQRLA |
2024-04-29 | Details |
|
|
Qwen 0.5B | "Qwen 0.5B: Bi-lingual LLM with 5B params & 32k context-length, optimized for text generation via SFT, RLHF, and continued pretraining." | 1B | fp16 q2 q3 q32 q4 q5 q6 q8 | 0.5 GB |
TQRLA TQ-LA TQRLA TQRLA TQRLA |
2024-04-29 | Details |
|
|
Zephyr 7B | Zephyr 7B: A 7 billion parameter, 2048 context-length model by Hugging Face for monolingual chat applications. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 MIT MIT |
2024-04-29 | Details |
|
|
Zephyr 141B | Zephyr 8X22B: A 22 billion parameter model by Hugging Face for multilingual chat, coding, and mathematical reasoning tasks. | 141B | fp16 q2 q3 q4 q5 q6 q8 | 70.5 GB |
MIT Apache-2.0 Apache-2.0 |
2024-04-29 | Details |
|
|
Dolphin Llama3 8B | "Dolphin Llama3 8B: An 8 billion parameter model by Cognitive Computations, supporting English with context lengths up to 8K for instructional tasks." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-04-29 | Details |
|
|
Dolphin Llama3 70B | "Dolphin Llama3 70B by Cognitive Computations: A powerful, open-source, English-only LLM with 70 billion parameters and an 8k context window, designed for instructional tasks." | 71B | fp16 q2 q3 q4 q5 q6 q8 | 35.5 GB |
META-LLAMA-3-CCLA META-LLAMA-3-CCLA |
2024-04-29 | Details |
| No Logo | Wizardlm2 7B | Dreamgen's WizardLM2 7B: A 7 billion parameter, monolingual large language model designed for chat with a context window of 4096 tokens. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 Apache-2.0 |
2024-04-29 | Details |
| No Logo | Wizardlm2 8X22B | Dreamgen's WizardLM2 141B: A 141 billion parameter, monolingual large language model with a 4096 token context window, designed for advanced chat applications. | 141B | fp16 q2 q3 q4 q5 q6 q8 | 70.5 GB |
Apache-2.0 Apache-2.0 |
2024-04-29 | Details |
|
|
Dolphincoder 7B | Dolphincoder 7B: Community-maintained, 7 billion param. LLM for coding aid. Supports mono-lingual contexts up to 4k/2k. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
BigCode-Open-RAIL-M-v1 BigCode-Open-RAIL-M-v1 |
2024-04-29 | Details |
|
|
Dbrx 132B Instruct | "Dbrx 132B by Databricks: A powerful 132 billion parameter model with a 32K context window, designed for English language text completion tasks." | 132B | fp16 q2 q4 q8 | 66.0 GB |
DOML DOML |
2024-04-29 | Details |
| No Logo | Qwen 0.5B | "Qwen 0.5B: Alibaba's bilingual LLM with 0.5 billion params, optimized for text generation tasks like SFT and RLHF." | 1B | fp16 q2 q3 q32 q4 q5 q6 q8 | 0.5 GB |
TQRLA |
2024-04-29 | Details |
| No Logo | Qwen 1.8B | "Qwen 1.8B: Alibaba's bilingual LLM with 1.8 billion params, 32k context-length, excelling in text generation via SFT & RLHF." | 2B | fp16 q2 q3 q32 q4 q5 q6 q8 | 1.0 GB |
TQRLA TQRLA |
2024-04-29 | Details |
| No Logo | Qwen 110B | "Qwen 110B: Alibaba's 110 billion param LLM for bilingual text gen. Supports SFT & RLHF." | 111B | fp16 q2 q3 q32 q4 q5 q6 q8 | 55.5 GB |
TQ-LA |
2024-04-29 | Details |
| No Logo | Qwen 14B | "Qwen 14B: Alibaba's 14 billion param. bi-lingual LLM for text gen & completion." | 14B | fp16 q2 q3 q32 q4 q5 q6 q8 | 7.0 GB |
TQ-LA TQ-LA |
2024-04-29 | Details |
| No Logo | Mxbai Embed Large 335M | "Mxbai Embed Large 335M: A 335M parameter model by Mixedbread, designed for generating sentence embeddings in retrieval tasks. Mono-lingual support." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-03-29 | Details |
|
|
Dolphin Mistral 7B | Dolphin Mistral 7B: Open-source, 7 billion param. LLM for chat assistance. Supports English. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 Apache-2.0 |
2024-03-29 | Details |
| No Logo | Stable Code 3B | "Stable Code 3B: A 3 billion parameter, 16k context-length LLM by Stability AI. Ideal for fine-tuning in application-specific tasks." | 3B | q2 q3 q32 q4 q5 q6 q8 | 1.5 GB |
STABILITY-AI-NCRCLA STABILITY-AI-NCRCLA |
2024-03-29 | Details |
| No Logo | Stable Code 3B Instruct | Stable Code 3B Instruct by Stability AI: A 3 billion parameter, monolingual model for general-purpose coding tasks. | 3B | q2 q3 q32 q4 q5 q6 q8 | 1.5 GB |
STABILITY-AI-NCRCLA STABILITY-AI-NCRCLA |
2024-03-29 | Details |
|
|
Starling Lm 7B | Starling LM 7B: Open-source, 7 billion param. model by Berkeley-Nest for conversational AI. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 Apache-2.0 |
2024-03-29 | Details |
| No Logo | All Minilm 22M | "All-MiniLM 22M by Sentence Transformers, a compact model for efficient information retrieval." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-02-29 | Details |
| No Logo | All Minilm 33M | "All-MiniLM-33M by Sentence Transformers: A compact, monolingual model for efficient information retrieval." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-02-29 | Details |
|
|
Nomic Embed 137M | "Nomic Embed 137M: A 137M parameter, monolingual language model by Nomic AI. Ideal for retrieval-augmented generation tasks with an 8k context window." | 0B | fp16 | 0.0 GB |
Apache-2.0 |
2024-02-29 | Details |
| No Logo | Llava 7B | Llava 7B: A 7 billion parameter, open-source large language model by Liuhaotian. Supports context lengths up to 32k for research in multimodal models and chatbot development. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 LLAMA-2-CLA Apache-2.0 LLAMA-2-CLA |
2024-01-29 | Details |
| No Logo | Llava 34B | Llava 34B: A 34 billion parameter large language model by Liuhaotian, supporting context lengths up to 3000 tokens. Ideal for research in multimodal models and chatbot development. | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
Apache-2.0 LLAMA-2-CLA Apache-2.0 |
2024-01-29 | Details |
| No Logo | Llava 13B | Llava 13B: A 13 billion parameter large language model by Liuhaotian, supporting multilingual contexts up to 32K tokens. Ideal for research in multimodal models and chatbot development. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
Apache-2.0 LLAMA-2-CLA LLAMA-2-CLA |
2024-01-29 | Details |
|
|
Phi 2.7B | "Phi 2.7B by Microsoft: A 7 billion parameter, monolingual LLM with a 2k context window, excelling in Q&A tasks." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
MIT MIT |
2024-01-29 | Details |
|
|
Openchat 7B | "Openchat 7B: A 7 billion parameter, bilingual large language model by Openchat. Ideal for coding tasks with an 8k context window." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 Apache-2.0 |
2024-01-29 | Details |
|
|
Tinydolphin 1.1B | Tinydolphin 1.1B: Open-source, 1 billion param LLM with 2k context length, ideal for monolingual tasks. | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
Apache-2.0 Apache-2.0 |
2024-01-29 | Details |
| No Logo | Wizardcoder 7B | "Wizardcoder 7B: A 7 billion parameter LLM by WizardLM, offering 2k context-length. Language support and main application field unspecified." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
MSRLT |
2024-01-29 | Details |
| No Logo | Wizardcoder 33B | "Wizardcoder 33B: 33 billion param, 2k context-length LLM by WizardLM. Mono-lingual, excels in code gen & completion." | 33B | fp16 q2 q3 q4 q5 q6 q8 | 16.5 GB |
MSRLT MSRLT |
2024-01-29 | Details |
| No Logo | Wizardcoder 34B | Wizardcoder 34B by WizardLM: 34 billion param. model for code gen., supports mono-lingual contexts up to 32k. | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
MSRLT |
2024-01-29 | Details |
| No Logo | Wizardcoder 13B | Wizardcoder 13B by WizardLM: 13B param. LLM for efficient code gen., supports mono-lingual models with context lengths up to 32k. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
MSRLT |
2024-01-29 | Details |
|
Sqlcoder 7B | Sqlcoder 7B by Defog.Ai: 7B params, 4k context, generates SQL from text. Mono-lingual. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
CC-BY-SA-4.0 |
2024-01-29 | Details |
|
|
Sqlcoder 15B | Sqlcoder 15B by Defog.Ai: 15B params, 4k context, generates SQL from text. Mono-lingual. | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
CC-BY-SA-4.0 CC-BY-SA-4.0 |
2024-01-29 | Details |
|
|
Sqlcoder 70B | Sqlcoder 70B by Defog.AI: 70B params, 4k context. Understands SQL databases for non-tech users. | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
CC-BY-SA-4.0 |
2024-01-29 | Details |
|
|
Llama Pro 8B | Llama Pro 8B: Arc Lab & Tencent's 8 billion param, 2k context-length model for general language tasks. | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
LLAMA-2-CLA |
2024-01-29 | Details |
|
|
Llama Pro 8B Instruct | Llama Pro 8B Instruct by Arc Lab & Tencent PCG. 8 billion params, 2k context length. Mono-lingual model for complex NLP tasks. | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
LLAMA-2-CLA |
2024-01-29 | Details |
|
|
Nexusraven 13B | "Nexusraven 13B: A 13 billion parameter LLM by Nexusflow, excelling in function calling with a 4k context window and monolingual language support." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
Nexusflowai-LT Nexusflowai-LT |
2024-01-29 | Details |
|
|
Duckdb Nsql 7B | The Duckdb Nsql 7B Text is an expansive language model developed by the company, Motherduck. This model boasts a capacity of 7 billion parameters, making it proficient in understanding and generating human-like text with remarkable accuracy and versatility. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2024-01-29 | Details | |
|
|
Megadolphin 120B | "Megadolphin 120B: A 120 billion parameter model by Cognitive Computations, offering 16k context-length for detailed conversations. Ideal for personalized advice and empathetic interactions in English." | 120B | fp16 q2 q3 q4 q5 q6 q8 | 60.0 GB |
LLAMA-2-CLA LLAMA-2-CLA |
2024-01-29 | Details |
| No Logo | Llama2 13B | "Llama2 13B by Meta Llama Enterprise: 13 billion param, 4k context-length LLM for assistant-style chat in English." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
LLAMA-2-AUP LLAMA-2-CLA |
2023-12-29 | Details |
| No Logo | Llama2 70B | "Llama2 70B by Meta Llama Enterprise: 70 billion param, 4k context-length LLM for assistant-style chat in English." | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
LLAMA-2-AUP LLAMA-2-CLA |
2023-12-29 | Details |
| No Logo | Llama2 7B | "Llama2 7B: Meta's 7 billion param, mono-lingual model for assistant-style chat." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-AUP LLAMA-2-CLA |
2023-12-29 | Details |
|
|
Deepseek Llm 67B Base | "DeepSeek LLM 67B Base: A 67 billion parameter, bilingual large language model by DeepSeek, designed for advanced NLP research." | 67B | fp16 q2 q3 q4 q5 q6 q8 | 33.5 GB |
DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Deepseek Llm 7B Base | "DeepSeek LLM 7B Base: Bi-lingual, 7B params, 4k context. Ideal for NLP research." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Nous Hermes2 34B | "Nous Hermes2 34B: Large language model by Nousresearch, featuring 34 billion parameters and a context-length of 4k. Ideal for developing Flask-based FTP servers." | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
Apache-2.0 |
2023-12-29 | Details |
| No Logo | Llama2 7B | "Llama2 7B: Meta's 7 billion param, 4k context-length model for assistant-style chats." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-AUP LLAMA-2-CLA LLAMA-2-AUP LLAMA-2-CLA |
2023-12-29 | Details |
| No Logo | Llama2 70B | "Llama2 70B by Meta Llama: 70 billion params, 4k context-length, mono-lingual, excels in assistant-style chat." | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
LLAMA-2-AUP LLAMA-2-CLA LLAMA-2-AUP LLAMA-2-CLA |
2023-12-29 | Details |
| No Logo | Llama2 13B | "Llama2 13B by Meta Llama: A 13 billion parameter, monolingual large language model with a 4096 token context window, designed for assistant-like chat applications." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
LLAMA-2-AUP LLAMA-2-CLA LLAMA-2-AUP LLAMA-2-CLA |
2023-12-29 | Details |
|
|
Deepseek Coder 1.3B Base | "Deepseek Coder 1.3B: Bi-lingual, 3B params, 16k context, excels in code completion." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Deepseek Coder 33B Base | "Deepseek Coder 33B: 33 billion param, 16k context-length LLM by Deepseek. Bi-lingual, excels in code completion." | 33B | fp16 q2 q3 q4 q5 q6 q8 | 16.5 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Deepseek Coder 6.7B Base | "Deepseek Coder 6.7B: 7B param, 16k context, bi-lingual, excels in code completion." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Deepseek Coder 1.3B Instruct | "Deepseek Coder 1.3B Instruct: Bi-lingual, 16k context-length model by Deepseek for advanced code completion." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Deepseek Coder 33B Instruct | "Deepseek Coder 33B Instruct: A powerful, bi-lingual language model by Deepseek. With 33 billion parameters and a context length of 16k, it excels in code completion and infilling tasks." | 33B | fp16 q2 q3 q4 q5 q6 q8 | 16.5 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Deepseek Coder 6.7B Instruct | "Deepseek Coder 7B Instruct: A 7 billion parameter, monolingual large language model by Deepseek. Ideal for generating code snippets from natural language prompts." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
DEEPSEEK-LICENSE DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Nous Hermes2 10.7B | Nous Hermes2 10.7B: A 7B parameter, monolingual LLM by Nousresearch, excelling in chat dialogue with a 2K context window. | 11B | fp16 q2 q3 q4 q5 q6 q8 | 5.5 GB |
Apache-2.0 Apache-2.0 |
2023-12-29 | Details |
|
|
Tinyllama 1.1B | TinyLlama 1.1B: University-maintained, 1B param, 2k context-length model for English chatbot applications. | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB | 2023-12-29 | Details | |
|
|
Openhermes 7B | OpenHermes 7B by Teknium: A 7 billion parameter, monolingual model with a 2k context window, excelling in general instruction following. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-12-29 | Details | |
|
|
Neural Chat 7B | "Neural Chat 7B by Intel: A 7 billion parameter, monolingual LLM with an 8k context window, excelling in language tasks." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-12-29 | Details | |
| No Logo | Wizard Math 7B | WizardMath 7B: A 7 billion parameter, 2048 context-length LLM by WizardLM for advanced mathematical reasoning. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
MSRLT MSRLT |
2023-12-29 | Details |
| No Logo | Wizard Math 70B | WizardMath 70B: A 70 billion parameter, monolingual large language model by WizardLM, excelling in mathematical reasoning with a context length of 4k. | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
MSRLT |
2023-12-29 | Details |
| No Logo | Wizard Math 13B | WizardMath 13B: A 13 billion parameter, monolingual large language model by WizardLM for text generation, with a context window of 2048 tokens. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
MSRLT |
2023-12-29 | Details |
|
Phind Codellama 34B | "Phind's Codellama 34B: A 34 billion parameter, monolingual LLM for coding tasks. Offers 4096 token context-length." | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB |
LLAMA-2-CLA LLAMA-2-CLA |
2023-12-29 | Details |
|
|
Dolphin Phi 2.7B | Dolphin Phi 2.7B: Open-source, 7B param LLM by Cognitive Computations for general chat, supports English with 2k context. | 3B | q2 q3 q4 q5 q6 q8 | 1.5 GB |
MSRLT MSRLT |
2023-12-29 | Details |
|
|
Solar 10.7B | Solar 10.7B: Upstage's 7B param, mono-lingual LLM, ideal for fine-tuning chatbots. | 11B | fp16 q2 q3 q4 q5 q6 q8 | 5.5 GB |
CC-BY-NC-4.0 |
2023-12-29 | Details |
|
|
Solar 10.7B Instruct | "Solar 10.7B Instruct by Upstage: A 7B parameter, monolingual model designed for single-turn conversations with a context length of 4k." | 11B | fp16 q2 q3 q4 q5 q6 q8 | 5.5 GB |
CC-BY-NC-4.0 |
2023-12-29 | Details |
|
|
Deepseek Llm 7B | DeepSeek LLM 7B: A 7 billion parameter, bilingual language model by DeepSeek AI, designed for text completion tasks with a context window of 2048 tokens. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
DEEPSEEK-LICENSE |
2023-12-29 | Details |
|
|
Deepseek Llm 67B | Deepseek Llm 67B: Bi-lingual, 67 billion param. model for text completion. | 67B | fp16 q2 q3 q4 q5 q6 q8 | 33.5 GB |
DEEPSEEK-LICENSE |
2023-12-29 | Details |
| No Logo | Bakllava 7B | Bakllava 7B: A 7 billion parameter, mono-lingual large language model by Llava Hugging Face, excelling in image-to-text generation with a context-length of 2k. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-12-29 | Details | |
| No Logo | Everythinglm 13B | "EverythingLM 13B: A 13 billion parameter, monolingual model by Totally-Not-An-Llm. Ideal for generating creative stories with a context-length of 16k." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
LLAMA-2-CLA LLAMA-2-CLA |
2023-12-29 | Details |
| No Logo | Stablelm Zephyr 3B | StableLM Zephyr 3B by Stability AI: A 3 billion parameter, English-only foundational model with a 4096 token context window, designed for application-specific fine-tuning. | 3B | q2 q3 q32 q4 q5 q6 q8 | 1.5 GB |
STABILITY-AI-NCRCLA STABILITY-AI-NCRCLA |
2023-12-29 | Details |
| No Logo | Magicoder 7B | "Magicoder 7B: ISE's 7 billion param, 1k context-length model for monolingual coding tasks." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-CLA |
2023-12-29 | Details |
|
|
Notux 8X7B | Notux 8X7B: Argilla's 7B param, multilingual LLM with 2k context-length. | 47B | fp16 q2 q3 q4 q5 q6 q8 | 23.5 GB |
MIT MIT |
2023-12-29 | Details |
|
|
Notus 7B | Notus 7B by Argilla: A 7 billion parameter, 2k context-length LLM for chat assistants. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
MIT MIT |
2023-12-29 | Details |
|
|
Alfred 40B | "Alfred 40B: Lightonio's 40 billion param, 1k context-length LLM for multilingual research in RLHF." | 42B | q4 q5 q8 | 21.0 GB |
Apache-2.0 |
2023-11-29 | Details |
|
|
Orca2 7B | "Orca2 7B: Microsoft's 7 billion parameter LLM, with 4k context-length, designed for monolingual research." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
MSRLT MSRLT |
2023-11-29 | Details |
|
|
Orca2 13B | "Orca2 13B by Microsoft: 13 billion params, 4k context-length. Mono-lingual model for research assessment." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
MSRLT MSRLT |
2023-11-29 | Details |
|
|
Meditron 7B | "Meditron 7B: A 7 billion parameter LLM by Epfllm for medical exam QA. Supports English with a context-length of 2k." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-CLA |
2023-11-29 | Details |
|
|
Meditron 70B | "Meditron 70B: A 70 billion parameter, monolingual LLM by Epfllm. Ideal for medical exam Q&A with a 4k context window." | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
LLAMA-2-CLA |
2023-11-29 | Details |
| No Logo | Goliath 120B | Goliath 120B: Alpindale's 120 billion parameter LLM. Supports English with context lengths up to 8K. | 118B | fp16 q2 q3 q4 q5 q6 q8 | 59.0 GB |
LLAMA-2-CLA |
2023-11-29 | Details |
| No Logo | Starcoder 15B | "Starcoder 15B by Bigcode: 15B params, 8k context. Mono-lingual, excels in code gen & completion." | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
BigCode-Open-RAIL-M-v1 |
2023-10-29 | Details |
|
|
Falcon 180B | "Falcon 180B: TII's 180 billion param, multi-lingual LLM for advanced research." | 180B | fp16 q4 q5 q8 | 90.0 GB |
FALCON-180B-TII-LICENSE-V10 |
2023-10-29 | Details |
|
|
Falcon 7B | "Falcon 7B: TII's 7 billion param, 2k context-length LLM for multilingual research." | 7B | fp16 q4 q5 q8 | 3.5 GB |
Apache-2.0 |
2023-10-29 | Details |
|
|
Falcon 180B | "Falcon 180B: TII's 180 billion param, 2k context-length LLM for multilingual research." | 180B | fp16 q4 q5 q8 | 90.0 GB |
FALCON-180B-TII-LICENSE-V10 FALCON-180B-TII-LICENSE-V10 |
2023-10-29 | Details |
|
|
Falcon 40B | "Falcon 40B: TII's 40 billion param, 2k context-length LLM for multilingual research." | 42B | fp16 q4 q5 q8 | 21.0 GB |
Apache-2.0 |
2023-10-29 | Details |
|
|
Falcon 7B Instruct | "Falcon 7B Instruct: TII's 7 billion param, bi-lingual LLM for text gen, with 2k context length." | 7B | fp16 q4 q5 q8 | 3.5 GB |
Apache-2.0 |
2023-10-29 | Details |
|
|
Falcon 40B Instruct | "Falcon 40B Instruct: TII's 40 billion param, 2k context-length LLM for bilingual chat & instruct tasks." | 42B | fp16 q4 q5 q8 | 21.0 GB |
Apache-2.0 |
2023-10-29 | Details |
| No Logo | Llama2 Uncensored 7B | Llama2 Uncensored 7B: Open-source, 7 billion param LLM for chatbots. 2k context window. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-AUP LLAMA-2-CLA LLAMA-2-AUP LLAMA-2-CLA |
2023-10-29 | Details |
| No Logo | Llama2 Uncensored 70B | Llama2 Uncensored 70B: 70 billion param, 2k context-length model by Llama2 7B Uncensored Chat. Mono-lingual, ideal for chatbot applications. | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
LLAMA-2-AUP LLAMA-2-CLA LLAMA-2-AUP LLAMA-2-CLA |
2023-10-29 | Details |
|
|
Mistral Openorca 7B | "Mistral Openorca 7B: 7 billion param LLM, optimized for consumer GPUs. Mono-lingual, context-length 8k." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
| No Logo | Orca Mini 3B | "Orca Mini 3B: A 3 billion parameter, mono-lingual language model by Psmathur-Orca. Generates text from prompts with a context-length of 1k." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
CC-BY-NC-SA-4.0 |
2023-10-29 | Details |
| No Logo | Orca Mini 70B | Orca Mini 70B: Community-maintained, 70 billion param. LLM for instruction following, supports English with 4k context. | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
CC-BY-NC-SA-4.0 |
2023-10-29 | Details |
| No Logo | Orca Mini 13B | "Orca Mini 13B: A 13 billion parameter, mono-lingual LLM by Psmathur-Orca. Offers 1024 token context-length for detailed text generation tasks." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
CC-BY-NC-SA-4.0 |
2023-10-29 | Details |
| No Logo | Orca Mini 7B | "Orca Mini 7B: A compact, 7 billion parameter model by Psmathur-Orca. Supports English, with a context window of 1024 tokens. Ideal for text generation tasks." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
CC-BY-NC-SA-4.0 |
2023-10-29 | Details |
| No Logo | Llama2 Chinese 7B | "Llama2 Chinese 7B: HFL's 7B param, 1.8k context-length model for Chinese text generation." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
| No Logo | Llama2 Chinese 13B | "Llama2 Chinese 13B: HFL's 13 billion param, monolingual Chinese model for content generation." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB | 2023-10-29 | Details | |
| No Logo | Vicuna 7B | Vicuna 7B: Open-source 7 billion param LLM by Large Model Systems. Supports English, ideal for research & chatbot development. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
| No Logo | Vicuna 33B | Vicuna 33B: Open-source, 33 billion param. LLM by Large Model Systems Org., supports English, ideal for research & chatbot development. | 32B | fp16 q2 q3 q4 q5 q6 q8 | 16.0 GB | 2023-10-29 | Details | |
| No Logo | Vicuna 13B | Vicuna 13B: Open-source, 13 billion param. LLM by Large Model Systems Org. Supports English, ideal for research & chatbot development. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB | 2023-10-29 | Details | |
|
|
Wizard Vicuna Uncensored 7B | Wizard Vicuna Uncensored 7B: Open-source, 7 billion param LLM by Cognitive Computations. Supports English, excels in conversation & instruction following with a 2k context window. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
|
|
Wizard Vicuna Uncensored 30B | Wizard Vicuna Uncensored 33B: Community-maintained, 33 billion param. LLM for open-ended conversations. | 32B | fp16 q2 q3 q4 q5 q6 q8 | 16.0 GB | 2023-10-29 | Details | |
|
|
Wizard Vicuna Uncensored 13B | Wizard Vicuna Uncensored 13B: Community-maintained, 13 billion param. LLM for general convo & task execution. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB | 2023-10-29 | Details | |
|
|
Starcoder 3B Base | "Starcoder 3B by Bigcodeproject: A 3 billion parameter large language model for code generation & completion. Supports monolingual contexts up to 16k/4k." | 3B | fp16 q2 q3 q4 q5 q6 q8 | 1.5 GB |
BigCode-Open-RAIL-M-v1 |
2023-10-29 | Details |
|
|
Starcoder 15B Base | "Starcoder 15B by Bigcodeproject: A 15 billion parameter large language model for generating code snippets across multiple languages. Offers context lengths of up to 16k tokens." | 16B | fp16 q2 q3 q4 q5 q6 q8 | 8.0 GB |
BigCode-Open-RAIL-M-v1 |
2023-10-29 | Details |
|
|
Starcoder 7B Base | "Starcoder 7B by Bigcodeproject: 7 billion param. LLM for coding tasks, supports 16k/4k context lengths, generates code snippets in multiple languages." | 8B | fp16 q2 q3 q4 q5 q6 q8 | 4.0 GB |
BigCode-Open-RAIL-M-v1 |
2023-10-29 | Details |
|
|
Starcoder 1B Base | "Starcoder 1B by Bigcodeproject: 1 billion param, 8k context-length model for monolingual technical coding assistance." | 1B | fp16 q2 q3 q4 q5 q6 q8 | 0.5 GB |
BigCode-Open-RAIL-M-v1 |
2023-10-29 | Details |
|
|
Nous Hermes 7B | Nous Hermes 7B: A 7 billion parameter, 2048 context-length LLM by Nousresearch for multilingual general tasks. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
|
|
Nous Hermes 13B | "Nous Hermes 13B: A 13 billion parameter, monolingual language model by Nousresearch. Ideal for creative text generation with a context window of 2048 tokens." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB | 2023-10-29 | Details | |
|
|
Nous Hermes 70B | "Nous Hermes 70B: A 70 billion parameter, monolingual language model by Nousresearch. Ideal for creative text generation with a context-length of 4k." | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB | 2023-10-29 | Details | |
|
|
Yarn Llama2 7B | "Yarn Llama2 7B by Nousresearch: 7 billion param, 64k context-length LLM for handling extensive data in diverse applications." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
|
|
Yarn Llama2 13B | "Yarn Llama2 13B by NousResearch: A powerful, open-source large language model with 13 billion parameters and a context length of 128k. Ideal for handling long context data in diverse applications." | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB | 2023-10-29 | Details | |
|
|
Xwinlm 7B | "XwinLM 7B: A 7 billion parameter, bi-lingual large language model by Xwin-LM. Ideal for NLP tasks with a context window of 4096 tokens." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-CLA |
2023-10-29 | Details |
|
|
Xwinlm 13B | XwinLM 13B: Community-maintained, 13 billion param. LLM for code gen, supports 2k context. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
LLAMA-2-CLA |
2023-10-29 | Details |
|
|
Xwinlm 70B | "XwinLM 70B: A 70 billion parameter, bi-lingual large language model by Xwin-LM. Ideal for NLP tasks with a context window of 4k." | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
LLAMA-2-CLA |
2023-10-29 | Details |
|
|
Samantha Mistral 7B | Samantha Mistral 7B: Open-source, 7 billion param. LLM for conversation. Supports English, context window of 2048 tokens. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
|
|
Samantha Mistral 7B Instruct | Samantha Mistral 7B Instruct: Open-source, 7 billion param. LLM for general assistance. Community-maintained by Cognitive Computations. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
| No Logo | Wizardlm 70B | WizardLM 70B: A 70 billion parameter large language model by WizardLM, supporting English with context lengths up to 32K for executing complex instructions. | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
LLAMA-2-CLA |
2023-10-29 | Details |
| No Logo | Wizardlm 30B | "WizardLM 30B: A 30 billion parameter, monolingual large language model by WizardLM. Ideal for generating code solutions with a context-length of 2k." | 32B | fp16 q2 q3 q4 q5 q6 q8 | 16.0 GB |
LLAMA-2-CLA |
2023-10-29 | Details |
| No Logo | Wizardlm 13B | WizardLM 13B: A 13-billion parameter, monolingual language model by WizardLM, designed for text generation with a context length of 2048 tokens. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
LLAMA-2-CLA |
2023-10-29 | Details |
| No Logo | Wizardlm 7B | "WizardLM 7B: A 7 billion parameter, monolingual language model by WizardLM. Offers a context-length of up to 2k." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
LLAMA-2-CLA |
2023-10-29 | Details |
| No Logo | Stable Beluga 7B | Stable Beluga 7B: A 7 billion parameter, 2k context-length model by Stability AI for monolingual chatting. | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
STABLE-BELUGA-NON-COMMERCIAL-COMMUNITY-LICENSE STABLE-BELUGA-NON-COMMERCIAL-COMMUNITY-LICENSE |
2023-10-29 | Details |
| No Logo | Stable Beluga 70B | Stable Beluga 70B: A 70 billion parameter, monolingual large language model by Stability AI, designed for engaging chats with a context window of 2048 tokens. | 69B | fp16 q2 q3 q4 q5 q6 q8 | 34.5 GB |
STABLE-BELUGA-NON-COMMERCIAL-COMMUNITY-LICENSE STABLE-BELUGA-NON-COMMERCIAL-COMMUNITY-LICENSE |
2023-10-29 | Details |
| No Logo | Stable Beluga 13B | Stable Beluga 13B: A 13 billion parameter, monolingual large language model by Stability AI, designed for engaging chats with a context window of 2048 tokens. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
STABLE-BELUGA-NON-COMMERCIAL-COMMUNITY-LICENSE STABLE-BELUGA-NON-COMMERCIAL-COMMUNITY-LICENSE |
2023-10-29 | Details |
| No Logo | Medllama2 7B | "MedLlama2 7B: A 7 billion parameter model by Doctorgpt, offering extensive context-length for detailed conversations." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
|
|
Wizardlm Uncensored 13B | WizardLM Uncensored 13B: Community-maintained, 13-billion param. LLM for conversational AI, supports English with 2k context. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB |
LLAMA-2-CLA |
2023-10-29 | Details |
|
|
Yarn Mistral 7B | "Yarn Mistral 7B by Nousresearch: A powerful monolingual LLM with 7 billion parameters, supporting context lengths up to 128k tokens. Ideal for tasks requiring long context handling like document summarization and detailed Q&A." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB | 2023-10-29 | Details | |
|
Codeup 13B | Codeup 13B: Open-source, 13 billion param LLM by Deepse. Bi-lingual, 2k context, excels in code generation. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB | 2023-10-29 | Details | |
|
|
Codebooga 34B | Codebooga 34B: 34 billion param, 2k context-length LLM for monolingual code gen & execution. | 34B | fp16 q2 q3 q4 q5 q6 q8 | 17.0 GB | 2023-10-29 | Details | |
|
Mistrallite 7B | "Mistrallite 7B by AWS: 7B params, 32k/16k context, mono-lingual, excels in long-context tasks." | 7B | fp16 q2 q3 q4 q5 q6 q8 | 3.5 GB |
Apache-2.0 |
2023-10-29 | Details |
|
|
Wizard Vicuna 13B | Wizard Vicuna 13B: Community-maintained, 13 billion param. LLM for general convo & instruction following. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB | 2023-10-29 | Details | |
|
|
Open Orca Platypus2 13B | Open Orca Platypus2 13B: Community-maintained, 13 billion param LLM. Supports English, excels in instruction & logical tasks. | 13B | fp16 q2 q3 q4 q5 q6 q8 | 6.5 GB | 2023-10-29 | Details |