AImodelsComparison.com 1,054 models
Menu
Models Best of Trending Can I run it? Methodology About
1,054 models · refreshed nightly

Models that run on RTX 4090 · 24 GB

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 410 text-to-image 153 ASR 92 image-text-to-text 89 sentence-similarity 61 feature-extraction 47 translation 47 text-classification 42 image-classification 34 token-classification 27 text-to-speech 12 audio-classification 11 object-detection 11 summarization 9 image-segmentation 9
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
$
719 models
#ModelParamsContextCommercial use30dOn RTX 4090 · 24 GB
51 Qwen3.8-27B-iMatrix-NVFP4-MTP-GGUF cdiamond · image-text-to-text — — ✓ apache-2.0 4.0 M 19.3 GB
52 Qwen3-1.7B Qwen · text-generation 2.0 B 41 K ✓ apache-2.0 3.8 M 5.3 GB
53 Ornith-1.0-35B-GGUF deepreinforce-ai · text-generation — — ✓ mit 3.8 M 23.8 GB
54 Qwen3-VL-4B-Instruct Qwen · image-text-to-text 4.4 B — ✓ apache-2.0 3.7 M 10.9 GB
55 distilbert-base-uncased-finetuned-sst-2-english distilbert · text-classification 70 M 512 ✓ apache-2.0 3.7 M 0.8 GB
56 Kimi-K3-DSpark RadixArk · text-generation 2.3 B 1.0 M unknown 3.6 M 5.8 GB
57 Qwen2.5-7B-Instruct-AWQ Qwen · text-generation 7.6 B 33 K ✓ apache-2.0 3.5 M 7.8 GB
58 NVIDIA-Nemotron-3-Nano-4B-BF16 nvidia · text-generation 4.0 B 262 K other 3.5 M 9.8 GB
59 pythia-160m EleutherAI · text-generation 210 M 2 K ✓ apache-2.0 3.5 M 0.9 GB
60 bert-large-cased-finetuned-conll03-english dbmdz · token-classification 330 M 512 unknown 3.4 M 2.0 GB
61 nsfw_image_detection Falconsai · image-classification 90 M — ✓ apache-2.0 3.4 M 0.9 GB
62 Ornith-1.0-9B-GGUF ornith-ai · text-generation — — ✓ mit 3.3 M 6.7 GB
63 nomic-embed-text-v1 nomic-ai · sentence-similarity 140 M 8 K ✓ apache-2.0 3.2 M 1.1 GB
64 Qwen3-VL-2B-Instruct Qwen · image-text-to-text 2.1 B — ✓ apache-2.0 3.0 M 5.5 GB
65 Florence-2-base microsoft · image-text-to-text 230 M — ✓ mit 3.0 M 1.0 GB
66 all-MiniLM-L6-v2 Xenova · feature-extraction 20 M 512 ✓ apache-2.0 3.0 M 0.5 GB
67 gemma-3-1b-it google · text-generation · gated 1.0 B — ⚠ gemma 2.9 M 2.8 GB
68 whisper-small openai · ASR 240 M — ✓ apache-2.0 2.9 M 1.6 GB
69 Huihui-Qwen3.8-27B-abliterated-GGUF huihui-ai · image-text-to-text 27.0 B — ✓ apache-2.0 2.8 M 16.5 GB
70 bge-reranker-large BAAI · feature-extraction 560 M 512 ✓ mit 2.8 M 3.0 GB
71 chandra-ocr-2 datalab-to · image-text-to-text 5.3 B — ⚠ openrail 2.7 M 12.9 GB
72 all-distilroberta-v1 sentence-transformers · sentence-similarity 80 M 512 ✓ apache-2.0 2.7 M 0.9 GB
73 mms-300m-1130-forced-aligner MahmoudAshraf · ASR 320 M — ✗ cc-by-nc-4.0 2.7 M 1.9 GB
74 embeddinggemma-300m google · sentence-similarity · gated 300 M — ⚠ gemma 2.6 M 1.9 GB
75 Qwen2.5-Coder-7B-Instruct Qwen · text-generation 7.6 B 33 K ✓ apache-2.0 2.6 M 18.4 GB
76 indonesian-roberta-base-posp-tagger w11wo · token-classification 120 M 512 ✓ mit 2.6 M 1.1 GB
77 Qwen3-TTS-12Hz-1.7B-CustomVoice Qwen · text-to-speech 1.9 B — ✓ apache-2.0 2.6 M 5.8 GB
78 Qwen3-Embedding-8B Qwen · feature-extraction 7.6 B 41 K ✓ apache-2.0 2.6 M 18.3 GB
79 Qwen3-14B-AWQ Qwen · text-generation 14.8 B 41 K ✓ apache-2.0 2.6 M 13.7 GB
80 Qwen3-VL-8B-Instruct-FP8 Qwen · image-text-to-text 8.8 B — ✓ apache-2.0 2.5 M 13.5 GB
81 Qwen2.5-VL-3B-Instruct Qwen · image-text-to-text 3.8 B 128 K unknown 2.5 M 9.3 GB
82 bge-base-en-v1.5 Xenova · feature-extraction 110 M 512 ✓ mit 2.5 M 0.8 GB
83 wav2vec2-large-robust-24-ft-age-gender audeering · audio-classification 320 M — ✗ cc-by-nc-sa-4.0 2.5 M 1.9 GB
84 t5-base google-t5 · translation 220 M — ✓ apache-2.0 2.4 M 1.5 GB
85 bge-base-en BAAI · feature-extraction 110 M 512 ✓ mit 2.4 M 1.0 GB
86 Qwen3.5-0.8B Qwen · image-text-to-text 870 M — ✓ apache-2.0 2.4 M 2.6 GB
87 mxbai-embed-large-v1 mixedbread-ai · feature-extraction 340 M 512 ✓ apache-2.0 2.4 M 1.3 GB
88 Ornith-1.0-9B deepreinforce-ai · text-generation <0.1 M — ✓ mit 2.4 M 21.2 GB
89 DeepSeek-OCR deepseek-ai · image-text-to-text 3.3 B 8 K ✓ mit 2.3 M 8.3 GB
90 Qwen3-ASR-1.7B Qwen · ASR 2.4 B — ✓ apache-2.0 2.3 M 6.0 GB
91 Qwen3.8-27B-Uncensored-GGUF JonathanColetti · text-generation — — ✓ apache-2.0 2.3 M 1.5 GB
92 Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF HauhauCS · image-text-to-text — — ✓ apache-2.0 2.3 M 1.5 GB
93 Unlimited-OCR baidu · image-text-to-text 3.3 B 33 K ✓ mit 2.2 M 8.3 GB
94 w2v-bert-2.0 facebook · feature-extraction 580 M — ✓ mit 2.2 M 3.1 GB
95 Ornith-1.0-35B-GGUF ornith-ai · text-generation — — ✓ mit 2.1 M 23.8 GB
96 Qwen3-Embedding-4B Qwen · feature-extraction 4.0 B 41 K ✓ apache-2.0 2.1 M 10.0 GB
97 SmolLM2-135M HuggingFaceTB · text-generation 130 M 8 K ✓ apache-2.0 2.1 M 0.8 GB
98 distilgpt2 distilbert · text-generation 90 M — ✓ apache-2.0 2.1 M 0.9 GB
99 Qwen2.5-14B-Instruct-AWQ Qwen · text-generation 14.8 B 33 K ✓ apache-2.0 2.1 M 13.7 GB
100 Qwen2-VL-2B-Instruct Qwen · image-text-to-text 2.2 B 33 K ✓ apache-2.0 2.1 M 5.7 GB
VRAM figures are estimates for the smallest available quantization at 8K context — see /methodology. Downloads refresh nightly from the Hugging Face API.