Measured by open model pull

Open models

How often an open AI model is downloaded.

10k100k1MAug 16Aug 31Sep 16Oct 3similar peersdeepseek-ai/DeepSee…951kdatalab-to/chandra357kCMSManhattan/JiRack…964k
#NameGrowth
1deepseek-ai/DeepSeek-V4-Flash-Vision-Expimage text to text model53×
2datalab-to/chandraimage text to text model27×
3CMSManhattan/JiRackUltra_14btext generation model26×
4tencent/Hy3text generation model25×
5fastino/gliner2.5-multi-v1token classification model20×
6jinaai/jina-embeddings-v5-omni-smallfeature extraction model18×
7neroued/Qwen3.8-27B-NInferimage text to text model16×
8zai-org/GLM-5.3text generation model15×
9sarvamai/sarvam-30btext generation model15×
10ibm-granite/granite-4.2-8btext generation model12×
11zai-org/GLM-5.3-Flashimage text to text model12×
12Infomaniak-AI/vllm-translategemma-12b-itimage text to text model12×
13jinaai/jina-embeddings-v5-text-small-text-matchingsentence similarity model9.7×
14nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16text generation model9.0×
15Comfy-Org/TRELLIS.2comfy.org8.5×
16Comfy-Org/Pixal3Dcomfy.org8.0×
17jinaai/jina-embeddings-v3-hffeature extraction model7.4×
18trklou/audio.cpp7.0×
19depth-anything/DA3-GIANTdepth estimation model6.9×
20poolside/Laguna-XS.2text generation model6.8×
21Qwen/Qwen3.8-Flash-Nextimage text to text model6.8×
22allenai/Olmo-3-7B-Thinktext generation model6.3×
23axolotl-ai-co/tiny-qwen2-129maxolotl.ai5.9×
24Octen/Octen-Embedding-8Bsentence similarity model5.9×
25vibevoice/VibeVoice-1.5B-hftext to speech model5.9×

At least as big as the median of its peers · 30 days · Downloads, not users. Models released in the last year; copies and converted versions are left out.

Measured, not estimated · snapshot 2026-10-03 · Growth is shown only where the starting size is big enough to mean something.