Arena Hard
Arena-Hard-Auto is an automatic evaluation benchmark for instruction-tuned LLMs consisting of 500 challenging real-world prompts curated by BenchBuilder. It includes open-ended software engineering problems, mathematical questions, and creative writing tasks. The benchmark uses LLM-as-a-Judge methodology with GPT-4.1 and Gemini-2.5 as automatic judges to approximate human preference. Arena-Hard achieves 98.6% correlation with human preference rankings and provides 3x higher separation of model performances compared to MT-Bench, making it highly effective for distinguishing between models of similar quality.
Qwen3 235B A22B from Alibaba Cloud / Qwen Team currently leads the Arena Hard leaderboard with a score of 0.956 across 26 evaluated AI models.
Qwen3 235B A22B leads with 95.6%, followed by Qwen3 32B at 93.8% and Qwen3 30B A3B at 91.0%.
Progress Over Time
Interactive timeline showing model performance evolution on Arena Hard
Arena Hard Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 |  Alibaba Cloud / Qwen Team | 235B | — | — | ||
| 2 | Alibaba Cloud / Qwen Team | 33B | 128K | $0.10 / $0.30 | ||
| 3 | Alibaba Cloud / Qwen Team | 31B | 128K | $0.10 / $0.44 | ||
| 4 | 50B | — | — | |||
| 5 |  Mistral AI | 24B | — | — | ||
| 6 | Alibaba Cloud / Qwen Team | 73B | — | — | ||
| 7 |  Microsoft | 14B | — | — | ||
| 8 |  DeepSeek | 236B | — | — | ||
| 9 | Microsoft | 15B | — | — | ||
| 10 | Microsoft | 14B | — | — | ||
| 11 | Mistral AI | 8B | — | — | ||
| 12 |  AI21 Labs | 398B | — | — | ||
| 13 | Mistral AI | 119B | 256K | $0.15 / $0.60 | ||
| 14 | 8B | — | — | |||
| 14 | 8B | — | — | |||
| 16 | Mistral AI | 675B | — | — | ||
| 16 | Mistral AI | 14B | — | — | ||
| 18 | Alibaba Cloud / Qwen Team | 8B | — | — | ||
| 19 | Mistral AI | 8B | — | — | ||
| 20 | AI21 Labs | 52B | — | — | ||
| 21 | Mistral AI | 24B | — | — | ||
| 22 | Microsoft | 60B | — | — | ||
| 23 | Microsoft | 4B | — | — | ||
| 24 | Microsoft | 4B | — | — | ||
| 25 | Mistral AI | 3B | — | — | ||
| 26 | 7B | — | — |
FAQ
Common questions about Arena Hard.
More evaluations to explore
Related benchmarks in the same category
A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.
A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.
Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions