The Open Weights
LatestModelsLeaderboardsUpcomingCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models
  • Upcoming launches

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • RSS feed
  • Newsletter

© 2026 The Open Weights. An independent publication.

Aggregated by Claude · written with Gemini · curated by humans.

Leaderboard

Open LLM

Aggregate of contamination-resistant benchmarks — IFEval, BBH, MATH, GPQA, MUSR, MMLU-Pro — for open-weight language models. The final snapshot of Hugging Face's Open LLM Leaderboard.

Source: Open LLM Leaderboard · Updated Jun 17, 2026 · 25 models
LLM IntelligenceText-to-ImageImage EditingText-to-VideoImage-to-VideoText-to-SpeechOpen LLMMost DownloadedCommunity Favorites
#ModelAvg.IFEvalMATHGPQA
1
MaziyarPanahi/calme-3.2-instruct-78b
52.180.640.320.4
2
MaziyarPanahi/calme-3.1-instruct-78b
51.381.439.319.5
3
dfurman/CalmeRys-78B-Orpo-v0.1
51.281.640.620
4
MaziyarPanahi/calme-2.4-rys-78b
50.880.140.720.4
5
huihui-ai/Qwen2.5-72B-Instruct-abliterated
48.185.960.119.4
6
Qwen/Qwen2.5-72B-Instruct
Qwen · Alibaba
48.086.459.816.7
7
MaziyarPanahi/calme-2.1-qwen2.5-72b
47.986.659.115.1
8
newsbang/Homer-v1.0-Qwen2.5-72B
47.576.34922.1
9
Saxo/Linkbricks-Horizon-AI-Avengers-V1-32B
47.379.760.315
10
MaziyarPanahi/calme-2.2-qwen2.5-72b
47.284.858.914.5
11
fluently-lm/FluentlyLM-Prinum
47.280.95418.2
12
JungZoona/T3Q-Qwen2.5-14B-Instruct-1M-e3
47.173.228.622.3
13
JungZoona/T3Q-qwen2.5-14b-v1.0-e3
47.173.228.622.3
14
zetasepic/Qwen2.5-32B-Instruct-abliterated-v2
46.983.359.515.7
15
rubenroy/Gilgamesh-72B
46.884.943.819.2
16
Sakalti/ultiima-72B
46.871.453.521.9
17
CombinHorizon/zetasepic-abliteratedV2-Qwen2.5-32B-Inst-BaseMerge-TIES
46.883.358.515.7
18
maldv/Awqward2.5-32B-Instruct
46.782.562.312.1
19
raphgg/test-2.5-72B
46.784.441.118.6
20
shuttleai/shuttle-3
46.781.54621.6
21
Qwen/Qwen2.5-32B-Instruct
Qwen · Alibaba
46.683.562.511.7
22
mistralai/Mistral-Large-Instruct-2411
Mistral AI
46.58449.524.9
23
rombodawg/Rombos-LLM-V2.5-Qwen-72b
46.571.654.219.8
24
Saxo/Linkbricks-Horizon-AI-Avengers-V3-32B
46.482.561.811.7
25
zetasepic/Qwen2.5-72B-Instruct-abliterated
46.371.552.420.9

Leaderboard data is cached from Open LLM Leaderboard and refreshed every 12 hours. Scores shown are for reference; see the source for full methodology.