1,222 models · 1,222 open weights · 200 API-only · refreshed nightly
Open-weights models
Models whose weights are published on Hugging Face — download them and run them on your own hardware. Many are also served by hosted providers; those carry an "Also via API" badge with live prices.
| # | Model | Params | Context | Commercial use | 30d | Min VRAM |
|---|---|---|---|---|---|---|
| 801 | twitter-xlm-roberta-base-sentiment-multilingual | — | 512 | unknown | 220 K | — |
| 802 | gemma4-e4b-claims-comparison | 7.9 B | — | ⚠ gemma | 219 K | from 19.2 GB |
| 803 | automotive | 34.7 B | — | ✓ apache-2.0 | 219 K | from 29.0 GB |
| 804 | Parable-Qwen3-8B-Claude-Fable-5-GGUF | 8.0 B | 41 K | ✓ apache-2.0 | 219 K | from 7.2 GB |
| 805 | hubert-large-speech-emotion-recognition-russian-dusha-finetuned | 320 M | — | ✓ apache-2.0 | 218 K | from 1.9 GB |
| 806 | Llama-3.1-Nemotron-Nano-8B-v1 | 8.0 B | 131 K | other | 218 K | from 19.4 GB |
| 807 | Qwen2.5-3B-Instruct-GGUF | — | — | other | 217 K | from 2.0 GB |
| 808 | sarvam-30b | 32.2 B | 131 K | ✓ apache-2.0 | 217 K | from 146.8 GB |
| 809 | Qwen2.5-Coder-1.5B-Instruct-GGUF | 1.5 B | 33 K | ✓ apache-2.0 | 216 K | from 1.6 GB |
| 810 | toxic-bert | 110 M | 512 | ✓ apache-2.0 | 216 K | from 1.0 GB |
| 811 | oneformer_cityscapes_swin_large | — | — | ✓ mit | 216 K | — |
| 812 | Gemma-4-E4B-DECKARD-HERETIC-NVFP4 | 6.2 B | — | ⚠ gemma | 216 K | from 12.6 GB |
| 813 | TinyLLama-v0 | <0.1 M | 2 K | ✓ apache-2.0 | 215 K | from 0.5 GB |
| 814 | gpt-oss-120b-GGUF | 120.0 B | 131 K | ✓ apache-2.0 | 215 K | from 19.4 GB |
| 815 | qwen3-4b-base-dapo-v4 | 4.0 B | 33 K | ✓ apache-2.0 | 215 K | from 10.0 GB |
| 816 | Qwen2-0.5B | 490 M | 131 K | ✓ apache-2.0 | 215 K | from 1.7 GB |
| 817 | Mellum2.1-12B-A2.5B-Thinking-GGUF | 12.0 B | — | ✓ apache-2.0 | 215 K | from 11.5 GB |
| 818 | sqlcoder-7b-2 | 6.7 B | 16 K | cc-by-sa-4.0 | 214 K | from 6.8 GB |
| 819 | Parable-Qwen3-4B-Claude-Fable-5-GGUF | 4.0 B | 41 K | ✓ apache-2.0 | 214 K | from 3.8 GB |
| 820 | MiniCPM5-1B-Claude-Opus-Fable5-Thinking-GGUF | — | — | ✓ apache-2.0 | 214 K | from 1.3 GB |
| 821 | OpenMed-NER-ChemicalDetect-ModernMed-149M | 150 M | 8 K | ✓ apache-2.0 | 213 K | from 0.9 GB |
| 822 | Qwen3Guard-Gen-0.6B | 750 M | 33 K | ✓ apache-2.0 | 213 K | from 2.3 GB |
| 823 | long-t5-tglobal-base-sci-simplify | 248 M | — | ✓ apache-2.0 | 213 K | from 1.6 GB |
| 824 | SmolLM3-3B-Base | 3.1 B | 66 K | ✓ apache-2.0 | 213 K | from 7.7 GB |
| 825 | NVIDIA-Nemotron-Nano-9B-v2 | 8.9 B | 131 K | other | 213 K | from 21.4 GB |
| 826 | Qwen-Image-2.1-Turbo-FP8 | 7.1 B | — | other | 213 K | from 61.1 GB |
| 827 | Qwen3-30B-A3B-Instruct-2507-FP8 | 30.5 B | 262 K | ✓ apache-2.0 | 213 K | from 39.4 GB |
| 828 | OTel-2.0-LLM-31B-IT | 31.3 B | 262 K | ✓ apache-2.0 | 212 K | from 144.6 GB |
| 829 | OTel-LLM-27B-IT | 27.0 B | — | ✓ apache-2.0 | 212 K | from 64.0 GB |
| 830 | OpenMed-NER-PharmaDetect-BigMed-278M | 278 M | 512 | ✓ apache-2.0 | 212 K | from 1.2 GB |
| 831 | wildguard | 7.3 B | — | ✓ apache-2.0 | 211 K | from 17.5 GB |
| 832 | OpenMed-NER-BloodCancerDetect-TinyMed-65M | 65 M | 512 | ✓ apache-2.0 | 211 K | from 0.7 GB |
| 833 | Qwen3-235B-A22B-Instruct-2507-FP8 | 235.1 B | 262 K | ✓ apache-2.0 | 211 K | from 295.8 GB |
| 834 | Ornith-1.0-35B-AEON-Ultimate-Uncensored-BF16 | 35.1 B | — | ✓ mit | 211 K | from 83.0 GB |
| 835 | Qwen3.6-35B-A3B-NVFP4-MTP-GGUF | — | — | unknown | 211 K | from 23.0 GB |
| 836 | Qwen2-1.5B-Instruct | 1.5 B | 33 K | ✓ apache-2.0 | 210 K | from 4.1 GB |
| 837 | Qwen-Image-Lightning | 20.4 B | — | ✓ apache-2.0 | 210 K | from 75.6 GB |
| 838 | controlnet-openpose-sdxl-1.0 | 1.3 B | — | ✓ apache-2.0 | 210 K | from 6.2 GB |
| 839 | Qwen3-30B-A3B-GPTQ-Int4 | 30.5 B | 41 K | ✓ apache-2.0 | 208 K | from 23.7 GB |
| 840 | Hy3-GGUF | — | — | ✓ apache-2.0 | 208 K | from 101.4 GB |
| 841 | DeepSeek-R1-Distill-Llama-70B-GGUF | — | 131 K | ⚠ llama3.3 | 208 K | from 29.5 GB |
| 842 | Bielik-11B-v3.0-Instruct-awq | 11.3 B | 33 K | ✓ apache-2.0 | 208 K | from 9.0 GB |
| 843 | Qwen3.6-35B-A3B-DFlash | 390 M | 262 K | ✓ apache-2.0 | 208 K | from 1.4 GB |
| 844 | Mistral-7B-Instruct-v0.3-AWQ | 7.3 B | 33 K | ✓ apache-2.0 | 208 K | from 6.2 GB |
| 845 | The_GuageLLM_23M | 20 M | — | ✓ apache-2.0 | 207 K | from 0.6 GB |
| 846 | Qwen3Guard-Gen-4B | 4.4 B | 33 K | ✓ apache-2.0 | 207 K | from 10.9 GB |
| 847 | Spark-X2.5-4B-Heretic-jp-gguf | 4.0 B | — | ✓ apache-2.0 | 206 K | from 4.0 GB |
| 848 | Mistral-7B-Instruct-v0.2-AWQ | 7.2 B | 33 K | ✓ apache-2.0 | 205 K | from 6.2 GB |
| 849 | Muse-Glimmer-30B | 29.8 B | — | ✓ apache-2.0 | 205 K | from 70.5 GB |
| 850 | FLUX.1-dev-gguf | — | — | other | 205 K | from 4.9 GB |
VRAM figures are estimates for the smallest available quantization at 8K context — see /methodology. Downloads refresh nightly from the Hugging Face API; API prices (blended 3:1 input:output per 1M tokens) and Artificial Analysis indices from OpenRouter.