Parameters
20.9B / 3.6B active
MoE total / active
Context
131,072 tokens
native · extended 131,072
Modalities
text text
input → output
Benchmarks
#116/138
better than 16% of all scored models
Size class
#17/24
better than 29% of similar-size models
24h trend
+0.6%
Interest delta, 24h
License
Apache License 2.0
About
gpt-oss-20b (openai/gpt-oss-20b) is OpenAI's smaller open-weight model, released August 4, 2025 under Apache 2.0 alongside gpt-oss-120b - a 20.91B-parameter Mixture-of-Experts transformer with 3.61B activated parameters per token, a 131,072-token context, and text-only input/output. Its MoE weights are post-trained with MXFP4 quantization, letting the model run within 16GB of memory; all reported evals use that same quantization.
It shares the 120b feature set: configurable reasoning effort (low, medium, high) tuned to latency needs, fully exposed chain-of-thought (for debugging and trust, not for end users), native agentic capabilities (function calling, web browsing, Python code execution, Structured Outputs), and full fine-tunability. It was trained on a text-only dataset of trillions of tokens focused on STEM, coding and general knowledge (knowledge cutoff June 2024), with harmful content filtered via CBRN pre-training filters from GPT-4o; training required ~10x fewer H100-hours than gpt-oss-120b. Model card paper: arXiv 2508.10925.
Training Data Trained on a text-only dataset with trillions of tokens, with a focus on STEM, coding, and general knowledge. Knowledge cutoff: June 2024. Pre-training data filtered for harmful content using CBRN pre-training filters from GPT-4o. Training required ~10x fewer H100-hours than gpt-oss-120b.
Benchmark Scores
| Benchmark | Score | Date | Standing |
|---|---|---|---|
|
AIME 2024 (with tools)
stem_reasoning
|
100.00%
|
05.08.2025 |
#1/1
only tracked model
|
|
AIME 2025
stem_reasoning
|
88.21%
|
05.08.2025 |
#12/28
better than 56%
|
|
AIME 2025 (with tools)
stem_reasoning
|
100.00%
|
05.08.2025 |
#1/1
only tracked model
|
|
GPQA Diamond
stem_reasoning
|
56.89%
|
05.08.2025 |
#70/84
better than 17%
|
|
GPQA Diamond (with tools)
stem_reasoning
|
100.00%
|
05.08.2025 |
#1/1
only tracked model
|
|
Humanity's Last Exam
stem_reasoning
|
16.67%
|
05.08.2025 |
#58/71
better than 19%
|
|
HLE (with tools)
stem_reasoning
|
6.06%
|
05.08.2025 |
#53/54
better than 2%
|
|
SWE-bench Verified
coding_agent
|
69.15%
|
05.08.2025 |
#46/62
better than 26%
|
|
Tau-Bench Retail
general_agent
|
54.80
|
05.08.2025 |
#3/3
better than 0%
|
|
Tau-Bench Airline
general_agent
|
38.00
|
05.08.2025 |
#3/3
better than 0%
|
|
Aider Polyglot
coding_agent
|
34.20
|
05.08.2025 |
#2/2
better than 0%
|
|
AIME 2024
stem_reasoning
|
92.10
|
05.08.2025 |
#7/7
better than 0%
|
|
MMMLU
multilingual
|
33.74%
|
05.08.2025 |
#21/23
better than 9%
|
|
HealthBench
general_capabilities
|
42.50
|
05.08.2025 |
#6/6
better than 0%
|
|
HealthBench Hard
general_capabilities
|
10.80
|
05.08.2025 |
#2/2
better than 0%
|
|
HealthBench Consensus
general_capabilities
|
82.60
|
05.08.2025 |
#2/2
better than 0%
|
|
CodeForces
stem_reasoning
|
63.08%
|
05.08.2025 |
#7/18
better than 65%
|
|
CodeForces (with tools)
stem_reasoning
|
100.00%
|
05.08.2025 |
#1/1
only tracked model
|
|
MMLU
knowledge
|
76.26%
|
05.08.2025 |
#16/19
better than 17%
|
What is a good AIME 2024 (with tools) score? What is a good AIME 2025 score? What is a good AIME 2025 (with tools) score?
Hardware Requirements
Estimated VRAM per quantization at 4,096 tokens context — MoE: all experts must fit in VRAM, only 3.6B are active per token
| Quantization | Weights | KV Cache | Total VRAM | On your hardware |
|---|---|---|---|---|
| FP16 — 16-bit | 41.8 GB | 0.6 GB | 44.5 GB | Which GPUs can run it? Calculate on your GPU |
| INT8 — 8-bit | 20.9 GB | 0.3 GB | 22.2 GB | Which GPUs can run it? Calculate on your GPU |
| INT4 — 4-bit | 10.5 GB | 0.1 GB | 11.1 GB | Which GPUs can run it? Calculate on your GPU |
| FP4 — 4-bit | 10.5 GB | 0.1 GB | 11.1 GB | Which GPUs can run it? Calculate on your GPU |
Consumer GPUs that fit this model with a single card (INT4):
Architecture
Mixture-of-Experts (MoE) Transformer
- Parameters
- 20.9B
- Active params
- 3.6B
- Layers
- 24 12× Sliding Window Attn + 12× Full Attention
- Attention
- Hybrid 64:8 Hybrid Attention
- Experts
- 32 top-4
- Hidden
- 2880
- Head dim
- 64
- Vocabulary
- 201K
- Context
- 131K
Source: Hugging Face config.json · GptOssForCausalLM · exact layer pattern · model repo
Training Pipeline
-
1
pretraining
Pretraining
-
2
sft
Post-Training: Distillation and SFT
-
3
rl
Reinforcement Learning for Reasoning and Tool Use
-
4
other
Safety Training: Deliberative Alignment
Training & Evaluation Datasets
| Name | Role | Size | Modalities | Collection |
|---|---|---|---|---|
| Text-only pre-training corpus (STEM, coding, general knowledge) | pretraining | — | — |
Linked Resources
gpt-oss-120b & gpt-oss-20b Model Card
https://arxiv.org/abs/2508.10925
OpenAI gpt-oss - Reference implementations and inference tools
https://github.com/openai/gpt-oss
OpenAI harmony - Harmony chat format implementation
https://github.com/openai/harmony
Introducing gpt-oss - OpenAI Blog
https://openai.com/index/introducing-gpt-oss/
Try gpt-oss
https://gpt-oss.com/
gpt-oss Collection on HuggingFace
https://huggingface.co/collections/openai/gpt-oss
Trend Analysis
24h Change
+0.6%
7d Change
+5.3%
30d Change
+12.7%
Current
15,900,000 pulls
followers
+0.1%
likes
+0.0%
downloads_all_time
+0.2%
downloads
+0.1%
No comments yet
Be the first to share your thoughts!