Up to four models
Claude Opus 4.6 vs Claude Opus 4.5
On the index (GPT-4 = 100), Claude Opus 4.6 scores 165 and Claude Opus 4.5 152. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude Opus 4.6
165149–181
WeightsClosed
ReleasedFeb 2026
Price in / out$5.0 / $25
Context1M
Core tests taken7 of 9
Claude Opus 4.5
152136–168
WeightsClosed
ReleasedNov 2025
Price in / out$5.0 / $25
Context200K
Core tests taken5 of 9
Claude Opus 4.6 leads Claude Opus 4.5 by 13 points, ahead on 5 of 5 shared core tests, at a similar price.
Test by testthird-party results
Claude Opus 4.6Claude Opus 4.5
Every result as a table
| Test | Claude Opus 4.6 | Claude Opus 4.5 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 90.5% | 86.1% |
| Humanity's Last Exam | 34.4% | 25.2% |
| LiveBench Data Analysis | 69.9% | 74.4% |
| LiveBench IF | 63.3% | 62.5% |
| LiveBench Language | 83.3% | 81.3% |
| LiveBench Reasoning | 88.7% | 80.1% |
| GPQA Diamond (Vals AI) | 89.6% | 85.9% |
| MMLU Pro | 89.1% | 87.3% |
| MMMU Pro | 83.9% | 83.0% |
| CL-bench | 20.7% | 21.1% |
| EBR-bench | 12.7% | 14.3% |
| SimpleBench | 67.6% | 62.0% |
| SimpleQA Verified | 47.0% | 45.7% |
| Maths | ||
| FrontierMath (tiers 1-3) | 66.0% | 34.4% |
| AIME-style maths (OTIS mock) | 94.4% | 86.1% |
| LiveBench Mathematics | 89.3% | 90.4% |
| Chess puzzles | 21.0% | 14.0% |
| FrontierMath (2025 set) | 40.7% | 20.7% |
| FrontierMath tier 4 (2025 set) | 22.9% | 4.2% |
| FrontierMath (tier 4) | 26.8% | 4.9% |
| Coding | ||
| SWE-bench Verified | 78.7% | 76.6% |
| LiveBench Agentic Coding | 49.0% | 39.7% |
| LiveBench Coding | 78.2% | 79.7% |
| SWE-bench Pro (public set) | 51.9% | 45.9% |
| Terminal-Bench 2.0 (Terminus 2 runs) | 62.9% | 57.8% |
| LiveCodeBench | 84.7% | 83.7% |
| SWE-bench | 78.2% | 76.4% |
| Vibe Code Bench v1.1 | 57.6% | 20.6% |
| GSO-Bench | 41.2% | 26.5% |
| WeirdML | 78.0% | 63.7% |
| Agents | ||
| DTBench | 91.2% | 89.9% |
| DeepResearch Bench | 55.3% | 54.8% |
| Furniture assembly | 28.3% | 28.3% |
| LMCA | 55.8% | 44.5% |
| Mystery game puzzles | 25.0% | 22.0% |
| Professional work | ||
| CorpFin v2 | 67.0% | 65.1% |
| LegalBench | 85.3% | 84.6% |
| MedCode | 49.1% | 49.2% |
| MedScribe | 86.7% | 85.3% |
| MortgageTax | 68.5% | 68.7% |
| TaxEval v2 | 76.0% | 74.9% |
Task length · METR47 sec45 sec
Tests at least two of these models have taken. Bold rows are the core tests.