AImodelsComparison.com 1,226 open · 200 API
Menu
Models Best of Trending Can I run it? Methodology About
1,226 models · 1,226 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.

Type
All 1,426 Open weights (self-host) 1,226 API-only 200
Task
All text-generation 534 text-to-image 155 image-text-to-text 112 ASR 86 sentence-similarity 64 translation 56 text-classification 51 feature-extraction 48 image-classification 39 token-classification 21 text-to-speech 19 audio-classification 12 image-segmentation 11 summarization 8 object-detection 8 audio-text-to-text 1 any-to-any 1
Runs on
Any RTX 3060 · 12 GB RTX 4070 · 16 GB RTX 4090 · 24 GB 2 × 24 GB Mac M3 · 24 GB Mac M4 Max · 64 GB
Licence
Size
$
534 models
#ModelParamsContextCommercial use30dMin VRAM
101 Ornith-1.0-9B ornith-ai · text-generation Open weights <0.1 M — ✓ mit 792 K from 21.2 GB
102 Qwen3-Coder-30B-A3B-Instruct-AWQ-4bit cyankiwi · text-generation Open weights 30.0 B 262 K ✓ apache-2.0 780 K from 24.9 GB
103 Ornith-1.5-9B-NVFP4 ornith-ai · text-generation Open weights 6.7 B — ✓ mit 763 K from 11.2 GB
104 NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 nvidia · text-generation Open weights Also via API 31.6 B 262 K other 743 K from 77.6 GB
105 DeepSeek-V3-0324 deepseek-ai · text-generation Open weights Also via API 684.5 B 164 K ✓ mit 734 K from 860.6 GB
106 SmolLM3-3B HuggingFaceTB · text-generation Open weights 3.1 B 66 K ✓ apache-2.0 719 K from 7.7 GB
107 Qwen3-8B-Base Qwen · text-generation Open weights 8.2 B 33 K ✓ apache-2.0 718 K from 19.7 GB
108 Qwen3.8-27B-NVFP4-MTP-GGUF esatapedico · text-generation Open weights 27.0 B — ✓ apache-2.0 714 K from 41.0 GB
109 Qwen3-32B-AWQ Qwen · text-generation Open weights 32.8 B 41 K ✓ apache-2.0 707 K from 26.7 GB
110 Qwen2.5-Coder-7B-Instruct-AWQ Qwen · text-generation Open weights 7.6 B 33 K ✓ apache-2.0 697 K from 7.8 GB
111 llama-3.3-70b-instruct-awq casperhansen · text-generation Open weights 70.6 B 131 K ⚠ llama3.3 691 K from 54.8 GB
112 Sharp-Spark-X2.5-4B-GGUF peculiar-ragdoll · text-generation Open weights 4.0 B — ✓ apache-2.0 687 K from 1.1 GB
113 Ornith-1.0-397B-FP8 deepreinforce-ai · text-generation Open weights 397.0 B — ✓ mit 673 K from 505.7 GB
114 LFM2.5-2.6B-GGUF LiquidAI · text-generation Open weights 2.6 B 131 K other 652 K from 2.6 GB
115 Qwen2.5-1.5B Qwen · text-generation Open weights 1.5 B 131 K ✓ apache-2.0 646 K from 4.1 GB
116 Qwen3-0.6B-Base Qwen · text-generation Open weights 600 M 33 K ✓ apache-2.0 634 K from 1.9 GB
117 Qwen3-Coder-30B-A3B-Instruct Qwen · text-generation Open weights Also via API 30.5 B 262 K ✓ apache-2.0 629 K from 72.3 GB
118 GLM-5.1 zai-org · text-generation Open weights Also via API 753.9 B 203 K ✓ mit 626 K from 1,772.1 GB
119 Qwen3-4B-Thinking-2507 Qwen · text-generation Open weights 4.0 B 262 K ✓ apache-2.0 621 K from 10.0 GB
120 Llama-3.1-8B meta-llama · text-generation · gated Open weights 8.0 B — ⚠ llama3.1 611 K from 19.4 GB
121 granite-4.1-3b ibm-granite · text-generation Open weights 3.4 B 131 K ✓ apache-2.0 608 K from 8.5 GB
122 Laguna-M.1 poolside · text-generation Open weights 225.8 B 262 K ✓ apache-2.0 595 K from 531.1 GB
123 Laguna-XS.2 poolside · text-generation Open weights 33.4 B 262 K ✓ apache-2.0 593 K from 79.1 GB
124 NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 nvidia · text-generation Open weights 67.2 B 262 K other 588 K from 98.9 GB
125 MiniCPM-SALA-AWQ-8bit cyankiwi · text-generation Open weights 3.1 B 524 K ✓ apache-2.0 586 K from 12.7 GB
126 Qwen3-4B-GGUF Qwen · text-generation Open weights — — ✓ apache-2.0 585 K from 3.2 GB
127 Qwen2.5-7B Qwen · text-generation Open weights 7.6 B 131 K ✓ apache-2.0 579 K from 18.4 GB
128 Qwen3-4B-Base Qwen · text-generation Open weights 4.0 B 33 K ✓ apache-2.0 578 K from 10.0 GB
129 Qwen3-8B-GGUF Qwen · text-generation Open weights — — ✓ apache-2.0 569 K from 6.0 GB
130 Qwen3.8-Flash-Next-GGUF AtomicChat · text-generation Open weights 180.0 B — other 569 K from 28.5 GB
131 EXAONE-3.5-7.8B-Instruct-AWQ LGAI-EXAONE · text-generation Open weights 7.8 B 33 K other 564 K from 7.5 GB
132 Qwen3-1.7B-Base Qwen · text-generation Open weights 1.7 B 33 K ✓ apache-2.0 557 K from 4.5 GB
133 MiniCPM5-2B-MLX openbmb · text-generation Open weights 2.5 B 131 K ✓ apache-2.0 548 K from 2.4 GB
134 Llama-3.2-1B meta-llama · text-generation · gated Open weights 1.2 B — ⚠ llama3.2 547 K from 3.4 GB
135 Qwen2.5-Coder-32B-Instruct Qwen · text-generation Open weights Also via API 32.8 B 33 K ✓ apache-2.0 546 K from 77.5 GB
136 Qwen3.8-4B-Distill-GGUF empero-ai · text-generation Open weights — — ✓ apache-2.0 541 K from 3.6 GB
137 Qwen3.8-27B-NVFP4 nvidia · text-generation Open weights 27.0 B — ✓ apache-2.0 540 K from 28.7 GB
138 macbert4csc-base-chinese shibing624 · text-generation Open weights 100 M 512 ✓ apache-2.0 537 K from 1.0 GB
139 gpt-oss-20b-GGUF unsloth · text-generation Open weights — 131 K ✓ apache-2.0 530 K from 13.1 GB
140 Qwen3.8-27B-OBLITERATED OBLITERATUS · text-generation Open weights 27.8 B — ✓ apache-2.0 520 K from 5.7 GB
141 Ornith-1.5-35B-A3B-FP8 ornith-ai · text-generation Open weights 36.0 B — ✓ mit 519 K from 49.2 GB
142 Qwen3.8-9B-Distill-GGUF empero-ai · text-generation Open weights — — ✓ apache-2.0 515 K from 6.9 GB
143 Qwen3-Coder-Next Qwen · text-generation Open weights Also via API 79.7 B 262 K ✓ apache-2.0 511 K from 187.7 GB
144 Underdog-Saluki-27B-1.0 ConwayResearch · text-generation Open weights 27.0 B — ✓ apache-2.0 509 K from 5.2 GB
145 Qwen2.5-Coder-7B-Instruct-GPTQ-Int4 Qwen · text-generation Open weights 7.6 B 33 K ✓ apache-2.0 508 K from 7.8 GB
146 POCKET-35B-GGUF FINAL-Bench · text-generation Open weights 35.0 B — ✓ apache-2.0 504 K from 14.8 GB
147 Ternary-Bonsai-27B-gguf prism-ml · text-generation Open weights — — ✓ apache-2.0 502 K from 1.2 GB
148 pythia-14m EleutherAI · text-generation Open weights 10 M 2 K ✓ apache-2.0 501 K from 0.5 GB
149 Llama-3.2-3B-Instruct-FP8-dynamic RedHatAI · text-generation Open weights 3.6 B 131 K ⚠ llama3.2 500 K from 5.9 GB
150 Laguna-XS-2.1 poolside · text-generation Open weights Also via API 33.4 B 262 K openmdw-1.1 500 K from 79.1 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.