Up to four models
Claude Opus 4.8 vs Claude Opus 4.7
On the index (GPT-4 = 100), Claude Opus 4.8 scores 170 and Claude Opus 4.7 167. 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 4.8
170154–186
WeightsClosed
ReleasedMay 2026
Price in / out$5.0 / $25
Context1M
Core tests taken6 of 9
Claude Opus 4.7
167151–183
WeightsClosed
ReleasedApr 2026
Price in / out$5.0 / $25
Context1M
Core tests taken6 of 9
Claude Opus 4.8 and Claude Opus 4.7 are too close to call (3 points apart, within the uncertainty), ahead on 3 of 4 shared core tests, at a similar price.
Test by testthird-party results
Claude Opus 4.8Claude Opus 4.7
Every result as a table
| Test | Claude Opus 4.8 | Claude Opus 4.7 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 91.0% | 90.1% |
| LiveBench Data Analysis | 66.0% | 78.3% |
| LiveBench IF | 72.0% | 66.7% |
| LiveBench Language | 79.7% | 77.9% |
| LiveBench Reasoning | 89.2% | 87.2% |
| GPQA Diamond (Vals AI) | 92.4% | 90.1% |
| MMLU Pro | 89.6% | 89.9% |
| MMMU Pro | 86.6% | 85.5% |
| EBR-bench | 28.6% | 19.1% |
| SimpleBench | 64.8% | 61.7% |
| SimpleQA Verified | 53.0% | 51.7% |
| Maths | ||
| FrontierMath (tiers 1-3) | 80.0% | 70.2% |
| AIME-style maths (OTIS mock) | 98.3% | 97.8% |
| LiveBench Mathematics | 94.3% | 92.8% |
| Chess puzzles | 52.0% | 41.0% |
| FrontierMath (2025 set) | 47.2% | 43.8% |
| FrontierMath tier 4 (2025 set) | 31.3% | 22.9% |
| FrontierMath (tier 4) | 56.1% | 31.7% |
| Coding | ||
| LiveBench Agentic Coding | 50.5% | 50.7% |
| LiveBench Coding | 81.8% | 82.1% |
| Code Migration | 47.3% | 43.9% |
| LiveCodeBench | 87.8% | 85.1% |
| ProgramBench | 1.0% | 0.0% |
| SWE-bench | 88.6% | 82.0% |
| Vibe Code Bench v1.1 | 82.7% | 71.0% |
| GSO-Bench | 47.1% | 44.1% |
| MirrorCode | 65.6% | 43.3% |
| WeirdML | 82.9% | 76.4% |
| Agents | ||
| APEX-Agents | 48.9% | 49.2% |
| DTBench | 94.9% | 94.7% |
| Furniture assembly | 42.5% | 33.3% |
| LMCA | 57.5% | 52.2% |
| Mystery game puzzles | 36.0% | 28.0% |
| Professional work | ||
| Finance Agent (v2) | 53.9% | 51.5% |
| CorpFin v2 | 66.7% | 66.1% |
| EMB | 69.4% | 63.7% |
| LegalBench | 83.6% | 85.3% |
| Legal Research Bench | 43.8% | 38.5% |
| MedCode | 53.2% | 54.9% |
| MedScribe | 85.8% | 83.0% |
| MortgageTax | 69.9% | 70.3% |
| Tax Agent Bench | 30.0% | 23.7% |
| TaxEval v2 | 75.6% | 75.3% |
Tests at least two of these models have taken. Bold rows are the core tests.