Up to four models
Muse Spark 1.3 vs GLM-5.3
On the index (GPT-4 = 100), Muse Spark 1.3 scores 172 and GLM-5.3 172. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Muse Spark 1.3172156–188
WeightsClosed
ReleasedSep 2026
Price in / out$1.25 / $4.25
Context1.05M
Core tests taken4 of 9
GLM-5.3172155–188
WeightsOpen
ReleasedAug 2026
Price in / out$0.077 / $1.4
Context1.05M
Core tests taken5 of 9
Muse Spark 1.3 and GLM-5.3 are too close to call (1 point apart, within the uncertainty), ahead on 3 of 4 shared core tests, at 4.9× the price.
Test by testthird-party results
Muse Spark 1.3GLM-5.3Every result as a table
| Test | Muse Spark 1.3 | GLM-5.3 |
|---|---|---|
| Reasoning & knowledge | ||
| LiveBench Data Analysis | 79.6% | 70.2% |
| LiveBench IF | 78.0% | 69.3% |
| LiveBench Language | 82.8% | 79.9% |
| LiveBench Reasoning | 89.6% | 85.8% |
| Maths | ||
| FrontierMath (tiers 1-3) | 74.4% | 68.8% |
| AIME-style maths (OTIS mock) | 99.2% | 91.1% |
| LiveBench Mathematics | 96.0% | 87.9% |
| ProofBench v1.1 | 55.0% | 49.0% |
| Chess puzzles | 37.0% | 37.0% |
| FrontierMath (tier 4) | 46.3% | 29.3% |
| Coding | ||
| Terminal-Bench 4.0 | 10.6% | 38.9% |
| LiveBench Agentic Coding | 64.1% | 60.9% |
| LiveBench Coding | 81.1% | 79.0% |
| Code Migration | 27.6% | 44.2% |
| CyberBench v1.1 | 69.4% | 72.1% |
| IOI | 43.9% | 68.4% |
| ProgramBench | 0.5% | 1.5% |
| Vibe Code Bench v1.1 | 82.9% | 78.1% |
| FrontierSWE | 25.8% | 30.2% |
| Agents | ||
| APEX-Agents | 58.6% | 56.6% |
| Terminal-Bench-Science | 4.3% | 5.7% |
| DTBench | 96.5% | 87.7% |
| LMCA | 53.9% | 55.5% |
| Mystery game puzzles | 25.0% | 33.0% |
| Professional work | ||
| Finance Agent (v2) | 58.9% | 55.8% |
| EMB | 62.7% | 56.3% |
| Legal Research Bench | 40.9% | 49.0% |
| Tax Agent Bench | 36.0% | 39.7% |
Tests at least two of these models have taken. Bold rows are the core tests.
