Up to four models
Muse Spark Open vs GLM-5.3-Flash
On the index (GPT-4 = 100), Muse Spark Open scores 161 and GLM-5.3-Flash 161. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
161145–177
WeightsOpen
ReleasedOct 2026
Price in / out$1.25 / $4.25
Context1.05M
Core tests taken6 of 9
161145–177
WeightsOpen
ReleasedAug 2026
Price in / out$0.15 / $0.50
Context1.05M
Core tests taken6 of 9
Muse Spark Open and GLM-5.3-Flash are too close to call (0 points apart, within the uncertainty), ahead on 2 of 6 shared core tests, at 8.4× the price.
Test by testthird-party results
Every result as a table
| Test | Muse Spark Open | GLM-5.3-Flash |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 88.6% | 90.1% |
| Maths | ||
| FrontierMath (tiers 1-3) | 69.8% | 55.8% |
| AIME-style maths (OTIS mock) | 98.1% | 93.9% |
| FrontierMath (tier 4) | 31.7% | 17.1% |
| Coding | ||
| Terminal-Bench 4.0 | 6.6% | 37.4% |
| Code Migration | 21.2% | 20.5% |
| Vibe Code Bench v1.1 | 76.2% | 30.8% |
| Agents | ||
| APEX-Agents | 33.4% | 52.8% |
| Professional work | ||
| Finance Agent (v2) | 61.3% | 57.9% |
| EMB | 53.7% | 55.9% |
| HLAB-Vals-1.1 | 25.0% | 10.4% |
| Legal Research Bench | 38.5% | 45.2% |
| SAGE | 85.8% | 89.3% |
| Tax Agent Bench | 26.3% | 31.6% |
| Novel problems | ||
| ARC-AGI-2 | 57.5% | 65.8% |
| ARC-AGI-1 | 89.0% | 91.0% |
Tests at least two of these models have taken. Bold rows are the core tests.