Home › Directory › LLM Benchmarks & Leaderboards
LLM Benchmarks & Leaderboards
Leaderboards, benchmarks and evaluation datasets for models.
JSON: /api/v1/directory.json (filter on category=llm-benchmarks)
MTEB Leaderboard
Massive Text Embedding Benchmark leaderboard for comparing embedding models.
FreeNo authHELM
Stanford's Holistic Evaluation of Language Models framework and leaderboards.
FreeNo authLMArena
Crowdsourced model comparison arena and leaderboard.
FreeNo authArtificial Analysis
Independent benchmarks of model quality, speed and price.
FreeAPI keyOpenRouter Rankings
Usage-based rankings of models across apps.
FreeNo authHugging Face Open LLM Leaderboard
Leaderboard for open-weight language models.
FreeNo authSWE-bench
Benchmark for software-engineering agents resolving real GitHub issues.
FreeNo authGAIA Benchmark
Benchmark for general AI assistants requiring tool use.
FreeNo authMMLU
Massive multitask language understanding benchmark.
FreeNo authHumanEval
Code generation benchmark.
FreeNo authGSM8K
Grade-school math benchmark.
FreeNo auth