AImodelsComparison.com 1,011 models
Menu
Models Best of Trending Can I run it? Methodology About
1,011 models · refreshed nightly

Text generation models

Every model in the catalog with its licence, estimated VRAM and daily-tracked downloads. Filters update the URL — share any view.

Task
All text-generation 400 text-to-image 150 ASR 89 image-text-to-text 82 sentence-similarity 55 feature-extraction 47 translation 44 text-classification 39 image-classification 32 token-classification 22 text-to-speech 16 audio-classification 12 image-segmentation 8 object-detection 8 summarization 7
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
$
400 models
#ModelParamsContextCommercial use30dMin VRAM
101 Qwen3-1.7B-Base Qwen · text-generation 1.7 B 33 K ✓ apache-2.0 961 K from 4.5 GB
102 Ornith-1.0-35B-FP8 deepreinforce-ai · text-generation 35.1 B — ✓ mit 945 K from 47.2 GB
103 Ornith-1.0-35B-FP8 ornith-ai · text-generation 35.1 B — ✓ mit 945 K from 47.2 GB
104 NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 nvidia · text-generation 31.6 B 262 K other 931 K from 74.7 GB
105 DeepSeek-V3-0324 deepseek-ai · text-generation 684.5 B 164 K ✓ mit 930 K from 860.6 GB
106 gpt-neox-20b EleutherAI · text-generation 20.7 B 2 K ✓ apache-2.0 927 K from 49.0 GB
107 MiniCPM5-1B openbmb · text-generation 1.1 B 131 K ✓ apache-2.0 926 K from 3.0 GB
108 Qwen2.5-7B Qwen · text-generation 7.6 B 131 K ✓ apache-2.0 924 K from 18.4 GB
109 MiniMax-M2.7 MiniMaxAI · text-generation 228.7 B 205 K other 924 K from 288.0 GB
110 Qwen3.6-35B-A3B-abliterated-v4 Bahushruth · text-generation 34.7 B 262 K ✓ apache-2.0 916 K from 82.0 GB
111 Qwen2.5-1.5B-Instruct-AWQ Qwen · text-generation 1.8 B 33 K ✓ apache-2.0 913 K from 2.5 GB
112 Phi-tiny-MoE-instruct microsoft · text-generation 3.8 B 4 K ✓ mit 907 K from 9.3 GB
113 NVIDIA-Nemotron-3-Super-120B-A12B-BF16 nvidia · text-generation 123.6 B 262 K other 899 K from 291.0 GB
114 Hy3-GGUF vcruz305 · text-generation — — ✓ apache-2.0 891 K from 52.9 GB
115 Qwen3-8B-FP8 Qwen · text-generation 8.2 B 41 K ✓ apache-2.0 876 K from 12.1 GB
116 deepseek-v4-gguf antirez · text-generation — — ✓ mit 875 K from 4.7 GB
117 SmolLM3-3B HuggingFaceTB · text-generation 3.1 B 66 K ✓ apache-2.0 872 K from 7.7 GB
118 Qwen3-Coder-Next-FP8-dynamic RedHatAI · text-generation 79.8 B 262 K ✓ apache-2.0 864 K from 102.4 GB
119 Qwen2.5-1.5B Qwen · text-generation 1.5 B 131 K ✓ apache-2.0 863 K from 4.1 GB
120 mamba-130m-hf state-spaces · text-generation 130 M — unknown 846 K from 1.1 GB
121 Qwen2.5-1.5B-quantized.w8a8 RedHatAI · text-generation 1.8 B 33 K ✓ apache-2.0 839 K from 3.2 GB
122 Kimi-K2.7-Code-NVFP4 nvidia · text-generation — — other 836 K from 655.2 GB
123 Llama-3.1-70B-Instruct meta-llama · text-generation · gated 70.6 B — ⚠ llama3.1 813 K from 166.3 GB
124 Llama-3.2-1B-Instruct-FP8 RedHatAI · text-generation 1.5 B 131 K ⚠ llama3.2 799 K from 3.0 GB
125 Kimi-K3-DSpark RadixArk · text-generation 2.3 B 1.0 M unknown 798 K from 5.8 GB
126 DeepSeek-R1-Distill-Qwen-32B deepseek-ai · text-generation 32.8 B 131 K ✓ mit 793 K from 77.5 GB
127 MiniMax-M2.5 MiniMaxAI · text-generation 228.7 B 197 K other 780 K from 288.0 GB
128 Llama-2-7b-hf meta-llama · text-generation · gated 6.7 B — ⚠ llama2 779 K from 16.3 GB
129 phi-2 microsoft · text-generation 2.8 B 2 K ✓ mit 771 K from 7.0 GB
130 pythia-70m-deduped EleutherAI · text-generation 100 M 2 K ✓ apache-2.0 769 K from 0.7 GB
131 DeepSeek-R1-Distill-Qwen-14B deepseek-ai · text-generation 14.8 B 131 K ✓ mit 766 K from 35.2 GB
132 NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4 nvidia · text-generation 18.2 B 262 K other 765 K from 24.5 GB
133 Ternary-Bonsai-27B-gguf prism-ml · text-generation — — ✓ apache-2.0 761 K from 1.2 GB
134 Llama-2-7b-chat-hf meta-llama · text-generation · gated 6.7 B — ⚠ llama2 753 K from 16.3 GB
135 Qwen3-0.6B-Base Qwen · text-generation 600 M 33 K ✓ apache-2.0 749 K from 1.9 GB
136 Qwen3.6-27B-Text-NVFP4-MTP sakamakismile · text-generation 16.7 B — ✓ apache-2.0 747 K from 24.6 GB
137 Qwen3-30B-A3B-Instruct-2507-AWQ-4bit cyankiwi · text-generation 5.3 B 262 K ✓ apache-2.0 744 K from 21.2 GB
138 NVIDIA-Nemotron-3-Nano-4B-BF16 nvidia · text-generation 4.0 B 262 K other 739 K from 9.8 GB
139 Qwen2.5-32B-Instruct-AWQ Qwen · text-generation 32.8 B 33 K ✓ apache-2.0 738 K from 26.7 GB
140 Qwen2.5-Coder-1.5B-Instruct Qwen · text-generation 1.5 B 33 K ✓ apache-2.0 728 K from 4.1 GB
141 Qwen3-4B-Base Qwen · text-generation 4.0 B 33 K ✓ apache-2.0 722 K from 10.0 GB
142 gemma-4-31B-it-NVFP4-turbo LilaRest · text-generation 32.5 B — ✓ apache-2.0 702 K from 26.6 GB
143 MobileLLaMA-1.4B-Chat mtgv · text-generation — 2 K ✓ apache-2.0 699 K —
144 Qwen3-235B-A22B Qwen · text-generation 235.1 B 41 K ✓ apache-2.0 695 K from 553.0 GB
145 Mistral-7B-v0.1 mistralai · text-generation 7.2 B 33 K ✓ apache-2.0 679 K from 17.5 GB
146 Ornith-1.0-397B-FP8 deepreinforce-ai · text-generation 397.0 B — ✓ mit 673 K from 505.7 GB
147 Ornith-1.0-397B-FP8 ornith-ai · text-generation 396.8 B — ✓ mit 673 K from 505.7 GB
148 phi-4 microsoft · text-generation 14.7 B 16 K ✓ mit 671 K from 34.9 GB
149 Qwen3-32B-AWQ Qwen · text-generation 32.8 B 41 K ✓ apache-2.0 651 K from 26.7 GB
150 bloom-560m bigscience · text-generation 560 M — ⚠ bigscience-bloom-rail-1.0 628 K from 1.8 GB
VRAM figures are estimates for the smallest available quantization at 8K context — see /methodology. Downloads refresh nightly from the Hugging Face API.