Up to four models
Claude Opus 4.8 vs Muse Spark 1.3
On the index (GPT-4 = 100), Claude Opus 4.8 scores 170 and Muse Spark 1.3 169. 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
169153–186
WeightsClosed
ReleasedSep 2026
Price in / out$1.25 / $4.25
Context1.05M
Core tests taken4 of 9
Claude Opus 4.8 and Muse Spark 1.3 are too close to call (1 point apart, within the uncertainty), ahead on 2 of 4 shared core tests, at 5.0× the price.
Test by testthird-party results
Claude Opus 4.8
Muse Spark 1.3
Every result as a table
| Test | Claude Opus 4.8 | Muse Spark 1.3 |
|---|---|---|
| Reasoning & knowledge | ||
| LiveBench Data Analysis | 66.0% | 79.6% |
| LiveBench IF | 72.0% | 78.0% |
| LiveBench Language | 79.7% | 82.8% |
| LiveBench Reasoning | 89.2% | 89.6% |
| GeoBench | 81.0% | 86.0% |
| SimpleBench | 64.8% | 81.8% |
| Maths | ||
| FrontierMath (tiers 1-3) | 80.0% | 74.4% |
| AIME-style maths (OTIS mock) | 98.3% | 99.2% |
| LiveBench Mathematics | 94.3% | 96.0% |
| Chess puzzles | 52.0% | 37.0% |
| FrontierMath (tier 4) | 56.1% | 46.3% |
| Coding | ||
| Terminal-Bench 4.0 | 23.2% | 10.6% |
| LiveBench Agentic Coding | 50.5% | 64.1% |
| LiveBench Coding | 81.8% | 81.1% |
| Code Migration | 47.3% | 27.6% |
| ProgramBench | 1.0% | 0.5% |
| Vibe Code Bench v1.1 | 82.7% | 82.9% |
| MirrorCode | 65.6% | 54.4% |
| Agents | ||
| APEX-Agents | 48.9% | 58.6% |
| Terminal-Bench-Science | 4.3% | 4.3% |
| DTBench | 94.9% | 96.5% |
| LMCA | 57.5% | 53.9% |
| Mystery game puzzles | 36.0% | 25.0% |
| Professional work | ||
| Finance Agent (v2) | 53.9% | 58.9% |
| EMB | 69.4% | 62.7% |
| Legal Research Bench | 43.8% | 40.9% |
| Tax Agent Bench | 30.0% | 37.9% |
Tests at least two of these models have taken. Bold rows are the core tests.