Exact model configurations ranked by the latest official LiveBench release: one benchmark's view of capability.
Model rank
Model adoption signals
12 models ranked by hugging Face downloads (30 days).
253.38M downloads
all-MiniLM-L6-v2 ranks first
Hugging Face downloads (30 days)
12
Models ranked
Jul 27
Snapshot recorded
2026
All 12 models by hugging Face downloads (30 days)
all-MiniLM-L6-v2 leads at 253.38M downloads; the middle of the board sits at 9.6M.
- Ahead of second place
- 224.9M
- Held by the top three
- 85%
- Middle of the board
- 9.6M
- Below half the leader
- 11 of 12
Rank, name and hugging Face downloads (30 days)Middle 9.6M
- 1all-MiniLM-L6-v2253.38M downloads
- 2Qwen3 0.6B28.51M downloads
- 3Qwen3 8B16.76M downloads
- 4Gemma 4 26B A4B IT13.23M downloads
- 5Qwen3.5 9B10.94M downloads
- 6Llama 3.2 1B Instruct10.26M downloads
- 7DeepSeek R18.93M downloads
- 8Whisper Large v3 Turbo8.38M downloads
- 9Stable Diffusion XL Base 1.01.47M downloads
- 10FLUX.1-dev557.39K downloads
- 11Phi-4 Mini Instruct412.52K downloads
- 12Mistral Small 3.1 24B Instruct251.29K downloads
Hub downloads cover the trailing 30 days · 3 checks behind each row · 100% confidence across rows.
| Rank | Name | Hugging Face downloads (30 days) | Why |
|---|---|---|---|
| 01 | all-MiniLM-L6-v2 | 253.38M downloads | This rank |
| 02 | Qwen3 0.6B | 28.51M downloads | This rank |
| 03 | Qwen3 8B | 16.76M downloads | This rank |
| 04 | Gemma 4 26B A4B IT | 13.23M downloads | This rank |
| 05 | Qwen3.5 9B | 10.94M downloads | Showing |
| 06 | Llama 3.2 1B Instruct | 10.26M downloads | This rank |
| 07 | DeepSeek R1 | 8.93M downloads | This rank |
| 08 | Whisper Large v3 Turbo | 8.38M downloads | This rank |
| 09 | Stable Diffusion XL Base 1.0 | 1.47M downloads | This rank |
| 10 | FLUX.1-dev | 557.39K downloads | This rank |
| 11 | Phi-4 Mini Instruct | 412.52K downloads | This rank |
| 12 | Mistral Small 3.1 24B Instruct | 251.29K downloads | This rank |
Why this rank
Qwen3.5 9B on Model Adoption Signals
Snapshot of Jul 27, 2026, 3:18 AM UTC · Link to this snapshot
Some explanation inputs are unavailable. Not published: the evidence frozen with this snapshot, comparable prior ranking.
Ranks a fixed, reviewed cohort of official model repositories by Hugging Face Hub downloads over the trailing 30 days.
5
Rank on this board
10,939,195 downloads
Score
3
Inputs behind the score
100% confidence · 1 of 1 components evidenced
What the score is built from
1 of 1 evidencedEvery component evidenced · 1 evidence record each
| Component | Contribution |
|---|---|
| Hugging face downloads 30d | 10,939,195 |
New in this snapshot
No change in rankno_prior_published_snapshot
No component moved since the previous snapshot.
Where each number comes from
3 public recordsDays since repository updateHugging face days since modified · observed Jul 27, 2026, 3:18 AM UTC147.1 days100% confidence
- Source
- Hugging Face Hub API
- Grade
- A
- Freshness
- fresh
- Locator
{"fields":["downloads","likes","lastModified","pipeline_tag","library_name","cardData.license","tags"],"endpoint":"/api/models/Qwen/Qwen3.5-9B","repositoryId":"Qwen/Qwen3.5-9B"}
Hugging Face downloads (30 days)Hugging face downloads 30d · observed Jul 27, 2026, 3:18 AM UTC10,939,195 downloads100% confidence
- Source
- Hugging Face Hub API
- Grade
- A
- Freshness
- fresh
- Locator
{"fields":["downloads","likes","lastModified","pipeline_tag","library_name","cardData.license","tags"],"endpoint":"/api/models/Qwen/Qwen3.5-9B","repositoryId":"Qwen/Qwen3.5-9B"}
Hugging Face likesHugging face likes total · observed Jul 27, 2026, 3:18 AM UTC1,756 likes100% confidence
- Source
- Hugging Face Hub API
- Grade
- A
- Freshness
- fresh
- Locator
{"fields":["downloads","likes","lastModified","pipeline_tag","library_name","cardData.license","tags"],"endpoint":"/api/models/Qwen/Qwen3.5-9B","repositoryId":"Qwen/Qwen3.5-9B"}
More in Model rank
Capability-first rankings across frontier and open models.
Official style-controlled text human-preference ratings with confidence bounds, variance, vote counts, and a frozen dataset revision.
Exact model-plus-inference configurations ranked by UC Berkeley BFCL V4 overall accuracy, with tool-use breakdowns, cost, latency, and source evidence.
See how often AI names you.
Enter your website. In about two minutes, Rank.ai asks ChatGPT, Claude and Gemini 12 questions your customers ask and grades how often they name you.
- Your grade out of 100How often AI names you, cites your site, and how high it ranks you.
- Who gets namedEvery competitor in the answers, most named first.
- The pages AI readsThe sources behind each answer.
- Three fixesWhat to fix first, with a brief for the first article.