AImodelsComparison.com 1,213 open · 210 API
Menu
Models Best of Trending Can I run it? Methodology MCP server About
1,423 models · 1,213 open weights · 210 API-only · refreshed nightly

Models that run on 2 × 24 GB

Every model in the catalog — open weights you can self-host and API-only models — with licence or API price, estimated VRAM and daily-tracked downloads. Filters update the URL — share any view.

Type
All 1,423 Open weights (self-host) 1,213 API-only 210
Task
All text-generation 735 text-to-image 165 image-text-to-text 110 ASR 82 sentence-similarity 63 text-classification 50 translation 50 feature-extraction 46 image-classification 41 text-to-speech 23 token-classification 22 summarization 9 audio-classification 9 image-segmentation 9 object-detection 7 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
$
871 models
#ModelParamsContextCommercial use30dOn 2 × 24 GB
451 llama2-embedding-1b-8k mesolitica · feature-extraction Open weights 947 M 33 K unknown 416 K 4.8 GB
452 saiga_llama3_8b IlyaGusev · text-generation Open weights 8.0 B 8 K other 415 K 19.4 GB
453 pythia-6.9b EleutherAI · text-generation Open weights 7.0 B 2 K ✓ apache-2.0 414 K 16.8 GB
454 EXAONE-3.5-7.8B-Instruct LGAI-EXAONE · text-generation Open weights 7.8 B 33 K other 411 K 36.1 GB
455 RMBG-2.0 briaai · image-segmentation · gated Open weights 221 M — other 408 K 1.5 GB
456 Qwen3-Coder-30B-A3B-Instruct-AWQ-4bit cyankiwi · text-generation Open weights 30.0 B 262 K ✓ apache-2.0 406 K 24.9 GB
457 Mistral-7B-v0.1 mistralai · text-generation Open weights 7.2 B 33 K ✓ apache-2.0 405 K 17.5 GB
458 locate-anything.cpp-gguf mudler · object-detection Open weights 3.8 B — other 405 K 6.3 GB
459 vit_base_patch16_224.augreg2_in21k_ft_in1k timm · image-classification Open weights 90 M — ✓ apache-2.0 402 K 0.9 GB
460 Qwen2.5-Coder-1.5B-Instruct Qwen · text-generation Open weights 1.5 B 33 K ✓ apache-2.0 402 K 4.1 GB
461 BioLORD-2023 FremyCompany · sentence-similarity Open weights 110 M 512 other 399 K 1.0 GB
462 LLM2Vec-Meta-Llama-3-8B-Instruct-mntp McGill-NLP · sentence-similarity Open weights 8.0 B 8 K ✓ mit 396 K 1.9 GB
463 mistral-7b-v0.3-bnb-4bit unsloth · text-generation Open weights 7.5 B 33 K ✓ apache-2.0 392 K 6.2 GB
464 Qwen2.5-Coder-1.5B Qwen · text-generation Open weights 1.5 B 33 K ✓ apache-2.0 387 K 4.1 GB
465 SmolLM3-3B HuggingFaceTB · text-generation Open weights 3.1 B 66 K ✓ apache-2.0 386 K 7.7 GB
466 Meta-Llama-3-8B-Instruct meta-llama · text-generation · gated Open weights 8.0 B — ⚠ llama3 386 K 19.4 GB
467 humanizer jialinyyzz · text-generation Open weights 12.0 B — ✓ apache-2.0 380 K 6.6 GB
468 Qwen-Image-2.1-Turbo-GGUF DogukanUrker · text-to-image Open weights — — other 379 K 3.1 GB
469 bloom-560m bigscience · text-generation Open weights 560 M — ⚠ bigscience-bloom-rail-1.0 378 K 1.8 GB
470 Qwen2.5-Coder-7B Qwen · text-generation Open weights 7.6 B 33 K ✓ apache-2.0 377 K 18.4 GB
471 Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF huihui-ai · text-generation Open weights — — ✓ mit 377 K 12.5 GB
472 Qwen2.5-Coder-7B-Instruct-GGUF Qwen · text-generation Open weights — — ✓ apache-2.0 376 K 3.8 GB
473 audiobox-aesthetics facebook · audio-classification Open weights 100 M — ✓ cc-by-4.0 375 K 1.0 GB
474 DeepSeek-R1-0528-Qwen3-8B deepseek-ai · text-generation Open weights 8.2 B 131 K ✓ mit 371 K 19.7 GB
475 deid_roberta_i2b2 obi · token-classification Open weights 350 M 512 ✓ mit 369 K 2.1 GB
476 Phi-3.5-mini-instruct microsoft · text-generation Open weights 3.8 B 131 K ✓ mit 367 K 9.5 GB
477 spark-s1-4b-v6-GGUF abhishek085 · text-classification Open weights 4.0 B — ✓ apache-2.0 367 K 4.9 GB
478 Qwen3-4B-Thinking-2507 Qwen · text-generation Open weights 4.0 B 262 K ✓ apache-2.0 364 K 10.0 GB
479 fullstop-punctuation-multilang-large oliverguhr · token-classification Open weights 560 M 512 ✓ mit 363 K 3.0 GB
480 gpt-oss-20b-GGUF ggml-org · text-generation Open weights 20.0 B 131 K ✓ apache-2.0 356 K 4.5 GB
481 roberta-base-openai-detector openai-community · text-classification Open weights 125 M 512 ✓ mit 353 K 1.1 GB
482 convnext_base.fb_in22k_ft_in1k timm · image-classification Open weights 90 M — ✓ apache-2.0 350 K 0.9 GB
483 wide_resnet101_2.tv_in1k timm · image-classification Open weights 127 M — ✓ bsd-3-clause 350 K 1.1 GB
484 Meta-Llama-3.1-8B-Instruct-GGUF bartowski · text-generation Open weights — — ⚠ llama3.1 344 K 3.7 GB
485 Hermes-3-Llama-3.1-8B NousResearch · text-generation Open weights 8.0 B 131 K ⚠ llama3 339 K 19.4 GB
486 Spark-X2.5-4B-GGUF abenzerps · text-generation Open weights 4.0 B — ✓ apache-2.0 338 K 3.7 GB
487 mask2former-swin-large-ade-semantic facebook · image-segmentation Open weights 220 M — other 336 K 1.5 GB
488 Pony_Diffusion_V6_XL LyliaEngine · text-to-image Open weights — — cdla-permissive-2.0 335 K 8.5 GB
489 resnet-18 microsoft · image-classification Open weights 12 M — ✓ apache-2.0 333 K 0.6 GB
490 Phi-4-mini-instruct microsoft · text-generation Open weights 3.8 B 131 K ✓ mit 332 K 9.5 GB
491 Ornith-1.5-9B-OBLITERATED OBLITERATUS · text-generation Open weights 9.7 B — ✓ mit 332 K 6.3 GB
492 Z-Image-Turbo unsloth · text-to-image Open weights 6.2 B — ✓ apache-2.0 329 K 37.5 GB
493 convnextv2_large.fcmae_ft_in22k_in1k_384 timm · image-classification Open weights 200 M — ✗ cc-by-nc-4.0 327 K 1.4 GB
494 Jan-v3.5-4B-gguf janhq · text-generation Open weights — — ✓ apache-2.0 327 K 2.8 GB
495 Ornith-1.5-9B-MLX-8bit ornith-ai · text-generation Open weights 9.0 B — unknown 326 K 12.3 GB
496 Ornith-1.5-9B-MLX ornith-ai · text-generation Open weights 9.0 B — unknown 326 K 21.5 GB
497 Flux-1-Schnell Comfy-Org · text-to-image Open weights 11.9 B — ✓ apache-2.0 325 K 47.4 GB
498 Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF Kujira · text-generation Open weights 27.0 B — ✓ apache-2.0 325 K 13.7 GB
499 inclusively-classification E-MIMIC · text-classification Open weights 110 M 512 ✗ cc-by-nc-sa-4.0 323 K 1.0 GB
500 DeepSeek-R1-Distill-Qwen-7B deepseek-ai · text-generation Open weights 7.6 B 131 K ✓ mit 322 K 18.4 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.