Anthropic
Claude Opus 4.5 benchmark results
32 results on 10 benchmarks from 5 sources.
Also listed as: claude-opus-4-5-20251101-high-32k · claude-opus-4-5-20251101 · claude-opus-4-5-20251101_16K · Claude Opus 4.5 (16k thinking) · claude-opus-4-5-20251101_32K · Claude Opus 4.5 (32k thinking) · Claude Opus 4.5 (no thinking) · claude-opus-4-5-20251101_unknown · Claude Opus 4.5 (unknown thinking) · claude-opus-4-5-20251101-thinking
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1474 | 41 | Overall, style control on; 37527 votes; 95% CI 1470–1478“claude-opus-4-5-20251101-high-32k” | Official leaderboard | Oct 8, 2026 |
| LMArena Text | 1470 | 47 | Overall, style control on; 72952 votes; 95% CI 1467–1473“claude-opus-4-5-20251101” | Official leaderboard | Oct 8, 2026 |
| LMArena WebDev | 1494 | 58 | Overall; 11034 votes; 95% CI 1486–1502“claude-opus-4-5-20251101-high-32k” | Official leaderboard | Oct 8, 2026 |
| LMArena WebDev | 1469 | 65 | Overall; 13426 votes; 95% CI 1462–1476“claude-opus-4-5-20251101” | Official leaderboard | Oct 8, 2026 |
| GPQA Diamond | 86.0% | 76 | Epoch AI run, accuracy; effort: 32k thinking | Independent run | Nov 24, 2025 |
| GPQA Diamond | 85.5% | 81 | Epoch AI run, accuracy; effort: 16k thinking | Independent run | Nov 25, 2025 |
| GPQA Diamond | 80.7% | 113 | Epoch AI run, accuracy; effort: no thinking | Independent run | Nov 24, 2025 |
| SWE-bench Verified | 76.7% | 9 | Epoch AI run (Epoch scaffold); effort: no thinking | Independent run | Feb 5, 2026 |
| SWE-bench Verified | 79.2% | 1 | agent: Sonar Foundation Agent“claude-opus-4-5” | Official leaderboard | Dec 5, 2025 |
| SWE-bench Verified | 79.2% | 1 | agent: live-SWE-agent“claude-opus-4-5-20251101” | Official leaderboard | Dec 15, 2025 |
| SWE-bench Verified | 74.4% | 12 | agent: mini-SWE-agent“claude-opus-4-5-20251101” | Official leaderboard | Nov 24, 2025 |
| AIME (OTIS Mock AIME 2024–2025) | 86.1% | 91 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: 32k thinking | Independent run | Nov 24, 2025 |
| AIME (OTIS Mock AIME 2024–2025) | 81.7% | 107 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: 16k thinking | Independent run | Nov 24, 2025 |
| AIME (OTIS Mock AIME 2024–2025) | 48.1% | 170 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: no thinking | Independent run | Nov 24, 2025 |
| Humanity's Last Exam | 25.2% | 17 | Compiled by Epoch AI; effort: unknown thinking | Third-party compilation via Epoch AI | Nov 24, 2025 |
| OSWorld (Verified) | 66.3% | 1 | OSWorld; agent: Claude Opus 4.5; effort: no thinking | Vendor-reported via Epoch AI | Nov 25, 2025 |
| τ²-bench | 84.0% | 1 | airline domain, pass^1; effort: high; submitted by Sierra; v1.0.1 | Official leaderboard | May 5, 2026 |
| τ²-bench | 24.7% | 20 | banking_knowledge domain, pass^1; effort: high; submitted by Sierra; v1.0.1 | Official leaderboard | May 5, 2026 |
| τ²-bench | 79.6% | 12 | retail domain, pass^1; effort: high; submitted by Sierra; v1.0.1 | Official leaderboard | May 5, 2026 |
| τ²-bench | 92.3% | 8 | telecom domain, pass^1; effort: high; submitted by Sierra; v1.0.1 | Official leaderboard | May 5, 2026 |
| Terminal-Bench 2.0 | 63.1% | 37 | Terminal-Bench 2.0; agent: Droid; effort: unknown thinking | Official leaderboard via Epoch AI | Nov 24, 2025 |
| Terminal-Bench 2.0 | 63.1% | 38 | Terminal-Bench 2.0; agent: Droid; effort: no thinking | Official leaderboard via Epoch AI | Dec 11, 2025 |
| Terminal-Bench 2.0 | 59.1% | 49 | Terminal-Bench 2.0; agent: Letta Code; effort: unknown thinking | Official leaderboard via Epoch AI | Nov 24, 2025 |
| Terminal-Bench 2.0 | 59.1% | 49 | Terminal-Bench 2.0; agent: Letta Code; effort: 128k thinking | Official leaderboard via Epoch AI | Dec 17, 2025 |
| Terminal-Bench 2.0 | 58.4% | 51 | Terminal-Bench 2.0; agent: Mux; effort: unknown thinking | Official leaderboard via Epoch AI | Nov 24, 2025 |
| Terminal-Bench 2.0 | 57.8% | 54 | Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking | Official leaderboard via Epoch AI | Nov 24, 2025 |
| Terminal-Bench 2.0 | 54.3% | 60 | Terminal-Bench 2.0; agent: Goose; effort: unknown thinking | Official leaderboard via Epoch AI | Nov 24, 2025 |
| Terminal-Bench 2.0 | 52.1% | 65 | Terminal-Bench 2.0; agent: Claude Code; effort: unknown thinking | Official leaderboard via Epoch AI | Nov 24, 2025 |
| Terminal-Bench 2.0 | 51.9% | 66 | Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking | Official leaderboard via Epoch AI | Nov 24, 2025 |
| Terminal-Bench 2.0 | 51.7% | 67 | Terminal-Bench 2.0; agent: OpenCode; effort: unknown thinking | Official leaderboard via Epoch AI | Nov 24, 2025 |
| BFCL v4 (function calling) | 77.5% | 1 | Overall accuracy; native function calling (FC)“Claude-Opus-4-5-20251101 (FC)” | Official leaderboard | Oct 17, 2026 |
| BFCL v4 (function calling) | 33.5% | 57 | Overall accuracy; prompt-based“Claude-Opus-4-5-20251101 (Prompt)” | Official leaderboard | Oct 17, 2026 |
FAQ
Where do these Claude Opus 4.5 scores come from?
From Epoch AI Benchmarking Hub, Berkeley Function Calling Leaderboard (BFCL), LMArena leaderboard dataset, SWE-bench leaderboards (swebench.com), and τ²-bench leaderboard submissions. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.