125 open models tracked live

What can you run on your GPU?

Pick your graphics card. See which open-source LLMs fit, how fast they run, and how much VRAM they need.

Open Hardware Compass View Top Models
125 Open Models 90 GPUs Tracked 6 New This Week 463 Benchmarks Tracked

Hardware Compass

Your GPU, your use case, your quantization. Find matching models in seconds.

90 GPUs in database

Optimized for task type

Lower = less VRAM

Pre-selected: GeForce RTX 4090 (24GB) — the most searched GPU

Top Open-Weights Models

Ranked by composite benchmark score & live trend momentum

Full Ranking
1 Steady

LFM3-VL-2B

LiquidAI · 1.9B · 62 benchmarks

2.7

composite score

1%
1GB Q4 LFM Open License v1.0
2 Steady

LFM3-Omni-2B

LiquidAI · 2.2B · 33 benchmarks

2.7

composite score

1%
1GB Q4 LFM Open License v1.0
3 Hot

LFM3-1.7B

LiquidAI · 1.7B · 29 benchmarks

2.7

composite score

34%
1GB Q4 LFM Open License v1.0
# Model Score VRAM (Q4) Trend License Compare
4 d1-2B

LiquidAI · 1.9B

 2.7 1GB Steady LFM Open License v1.0
5 Mistral Large 4

Mistral AI · 1.0T

 2.7 500GB Hot Mistral Ecosystem Open
6 Nanbeige4.3-33B

Nanbeige · 33.0B

 2.6 16GB Steady Apache License 2.0

Latest Releases

Newest open-source models, freshest first

All Models

Trending Models

Development across downloads, likes & followers — ranked by momentum

All Trends
LFM3-1.7B huggingface
Downloads +12.8%
Followers +1.7%
Momentum · 24h +16.5%

42,384

downloads +2 more

d1-2B huggingface
Downloads +18.6%
Followers +1.7%
Momentum · 24h +15.3%

5,047

downloads +2 more

LFM3-Omni-2B huggingface
Downloads +19.0%
Followers +1.7%
Momentum · 24h +14.5%

9,964

downloads +2 more

Compare Models Head-to-Head

Benchmark scores side by side

Saved Comparisons
vs
Popular comparisons
Qwen3.6-35B-A3B vs Qwen3.6-27B Qwen3.6-35B-A3B vs Ornith-1.0-35B

License Quick-Check

Find models you can actually use commercially

0 of 0 models match your criteria

Browse Matching Models