GPT-5 benchmark ranking on every AI leaderboard
closed weights · #43 overall, podium score 2.7 · #1 on 1 board · top 10 on 2 of 12 · released Aug 7, 2025
12 of 12 boards
| Rank | Board | Score | |||
|---|---|---|---|---|---|
| 7 |  MultiChallengeChat · as “gpt-5-thinking” | Chat | 63.2% | 29 | |
| 13 | HLEReasoning · as “gpt-5-2025-08-07” | Reasoning | 25.3% | 42 | |
| 30 |  Search ArenaAgents and tools · as “gpt-5-search” | Agents and tools | 1,133 | 33 | |
| 45 |  ARC-AGI-2Reasoning · as “GPT-5 (High)” | Reasoning | 9.9% | 78 | |
| 49 |  Epoch ECIReasoning · as “GPT-5” | Reasoning | 150.0 | 100 | |
| 52 | GPQA DiamondReasoning · as “gpt-5-2025-08-07_high” | Reasoning | 86.2% | 100 | |
| 54 | Vision ArenaVision · as “gpt-5-chat” | Vision | 1,225 | 100 | |
| 58 | WebDev ArenaCoding · as “gpt-5-medium” | Coding | 1,420 | 100 | |
| 76 | Text ArenaChat · as “gpt-5-high” | Chat | 1,435 | 100 | |
| 1 |  Aider PolyglotCoding · stale since 2025-10-03 · as “gpt-5 (high)” | Coding | 88.0% | 50 | |
| 16 |  MMLU-ProReasoning · stale since 2026-03-11 · as “GPT-5(high)” | Reasoning | 87.1% | 100 | |
| 16 |  SWE-bench VerifiedCoding · stale since 2026-02-26 · as “GPT 5 (2025-08-07) (medium)” | Coding | 65.0% | 39 |
Category ranks and more models
Sources: each leaderboard's own published results page, captured every day at 04:30 UTC. Scores exactly as each board published them; we run no evaluations. Logos via logo.dev; trademarks belong to their owners.