Updated 2026-11-10 · ranked by real download data

Models that fit in 12 GB

Ranked nightly from our download snapshots of the Hugging Face catalog. Every entry shows its licence and what hardware it realistically needs.

#	Model	Params	Context	Commercial use	30d	Min VRAM
01	all-MiniLM-L6-v2 +1 variant sentence-transformers · sentence-similarity	20 M	512	✓ apache-2.0	177.6 M	from 0.6 GB
02	bge-small-en-v1.5 BAAI · feature-extraction	30 M	512	✓ mit	47.6 M	from 0.7 GB
03	gpt2 openai-community · text-generation	140 M	—	✓ mit	31.8 M	from 1.1 GB
04	paraphrase-multilingual-MiniLM-L12-v2 sentence-transformers · sentence-similarity	120 M	512	✓ apache-2.0	27.1 M	from 1.0 GB
05	t5-small google-t5 · translation	60 M	—	✓ apache-2.0	21.6 M	from 0.8 GB
06	Qwen3-0.6B Qwen · text-generation	750 M	41 K	✓ apache-2.0	20.7 M	from 2.3 GB
07	all-mpnet-base-v2 sentence-transformers · sentence-similarity	110 M	512	✓ apache-2.0	20.5 M	from 1.0 GB
08	vit-base-patch16-224 google · image-classification	90 M	—	✓ apache-2.0	15.0 M	from 0.9 GB
09	bge-reranker-v2-m3 BAAI · text-classification	570 M	8 K	✓ apache-2.0	14.8 M	from 3.1 GB
10	mobilenetv3_small_100.lamb_in1k timm · image-classification	2.6 M	—	✓ apache-2.0	12.7 M	from 0.5 GB
11	Kokoro-82M hexgrad · text-to-speech	82 M	—	✓ apache-2.0	12.5 M	from 0.7 GB
12	nomic-embed-text-v1.5 nomic-ai · sentence-similarity	140 M	2 K	✓ apache-2.0	11.9 M	from 1.1 GB
13	Florence-2-base microsoft · image-text-to-text	230 M	—	✓ mit	11.7 M	from 1.0 GB
14	multilingual-e5-small intfloat · sentence-similarity	120 M	512	✓ mit	9.4 M	from 1.0 GB
15	Qwen3-Embedding-0.6B Qwen · feature-extraction	600 M	33 K	✓ apache-2.0	9.1 M	from 1.9 GB
16	paraphrase-multilingual-mpnet-base-v2 sentence-transformers · sentence-similarity	280 M	512	✓ apache-2.0	9.1 M	from 1.8 GB
17	audio.cpp-gguf audio-cpp · text-to-speech	—	—	other	8.9 M	from 1.6 GB
18	bge-base-en-v1.5 +1 variant BAAI · feature-extraction	110 M	512	✓ mit	8.3 M	from 1.0 GB
19	bge-large-en-v1.5 BAAI · feature-extraction	340 M	512	✓ mit	8.2 M	from 2.0 GB
20	Gender-Classifier-Mini prithivMLmods · image-classification	93 M	—	✓ apache-2.0	8.1 M	from 1.7 GB
21	clap-htsat-fused laion · audio-classification	150 M	—	✓ apache-2.0	7.3 M	from 1.2 GB
22	multilingual-e5-large intfloat · feature-extraction	560 M	512	✓ mit	7.1 M	from 3.0 GB
23	granite-embedding-small-english-r2 ibm-granite · feature-extraction	50 M	8 K	✓ apache-2.0	6.9 M	from 0.6 GB
24	Ornith-1.5-9B-GGUF ornith-ai · text-generation	—	—	✓ mit	5.8 M	from 6.9 GB
25	Qwen3.5-4B Qwen · image-text-to-text	4.7 B	—	✓ apache-2.0	5.3 M	from 11.5 GB

Membership and ranking refresh nightly after the snapshot run. VRAM is an estimate for the smallest quantization at 8K context — methodology.