Model components / Evaluation code

Open LLM Leaderboard

Hugging Face's flagship public leaderboard for comparing open LLMs on a fixed, reproducible benchmark suite run on a shared cluster. v2 (2024) raised difficulty with IFEval, BBH, MATH-L5, GPQA, MuSR, and MMLU-Pro. It evaluated 13,000+ models and drew 2M+ visitors before being archived.

ARCHIVED/retired (announced 2025-03-13); HF redirected users to the OpenEvals hub. HF Space was Apache-2.0; backend used lm-evaluation-harness (v1) and lighteval (v2). Listed for historical significance.

Openness

5 high confidence

5.0

license: Apache-2.0(HF-Space)
backend: lm-eval-harness/lighteval(open)
source: public

Open Apache-2.0 leaderboard built on open eval backends (now archived).

https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard recorded 2026-06-17
Apache-2.0 Space, ~14k likes

Adoption

4 high confidence

4.0

ARCHIVED 2025-03-13. At peak: 13,000+ models evaluated, 2M+ unique visitors - historically the most influential open-model leaderboard.

https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard/discussions/1135 recorded 2026-06-17
retirement announcement, 2025-03-13

Capability

4 high confidence

4.0

Historically the most visible open-model leaderboard; static rather than human-preference; now archived.

https://huggingface.co/docs/leaderboards/en/open_llm_leaderboard/archive recorded 2026-06-17
v2 benchmark suite, 2M visitors, archive

Unchanged since 2026-06-17 (last edited, not re-checked)