Leaderboard
The most-downloaded open-weight models on Hugging Face over the last 30 days — the community's working set, straight from the Hub and refreshed every 12 hours.
| # | Model | Downloads | Likes | Task |
|---|---|---|---|---|
| 1 | sentence-transformers/all-MiniLM-L6-v2 | 257.6M | 5.2K | sentence similarity |
| 2 | google-bert/bert-base-uncased | 114.9M | 2.7K | fill mask |
| 3 | cross-encoder/ms-marco-MiniLM-L6-v2 | 89.1M | 293 | text ranking |
| 4 | BAAI/bge-small-en-v1.5 BAAI | 72.5M | 520 | feature extraction |
| 5 | sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 | 60M | 1.3K | sentence similarity |
| 6 | google/electra-base-discriminator Google DeepMind | 55.8M | 148 | — |
| 7 | lpiccinelli/unidepth-v2-vitl14 | 36.2M | 47 | — |
| 8 | BAAI/bge-m3 BAAI | 35.2M | 3.3K | sentence similarity |
| 9 | Qwen/Qwen3-0.6B Qwen · Alibaba | 29M | 1.5K | text generation |
| 10 | amazon/chronos-2 | 28.4M | 388 | time series forecasting |
| 11 | sentence-transformers/all-mpnet-base-v2 | 25.8M | 1.3K | sentence similarity |
| 12 | google-t5/t5-small | 25.3M | 588 | translation |
| 13 | openai/clip-vit-base-patch32 | 22.9M | 992 | zero shot image classification |
| 14 | FacebookAI/xlm-roberta-base | 21.3M | 877 | fill mask |
| 15 | BAAI/bge-reranker-v2-m3 BAAI | 19.2M | 1.1K | text classification |
| 16 | timm/mobilenetv3_small_100.lamb_in1k | 18.5M | 99 | image classification |
| 17 | facebook/opt-125m | 17.9M | 286 | text generation |
| 18 | Qwen/Qwen3-8B Qwen · Alibaba | 15.9M | 1.3K | text generation |
| 19 | intfloat/multilingual-e5-small | 15.6M | 378 | sentence similarity |
| 20 | nomic-ai/nomic-embed-text-v1.5 | 15.5M | 883 | sentence similarity |
| 21 | trl-internal-testing/tiny-Qwen2ForCausalLM-2.5 | 15.4M | 18 | text generation |
| 22 | Qwen/Qwen2.5-1.5B-Instruct Qwen · Alibaba | 14M | 788 | text generation |
| 23 | autogluon/chronos-bolt-small | 13.7M | 58 | time series forecasting |
| 24 | openai-community/gpt2 | 13.6M | 3.4K | text generation |
| 25 | BAAI/bge-large-en-v1.5 BAAI | 13.4M | 707 | feature extraction |
Leaderboard data is cached from Hugging Face Hub and refreshed every 12 hours. Scores shown are for reference; see the source for full methodology.