Up to four models
Muse Ember vs Gemini 3.8 Flash
On the index (GPT-4 = 100), Muse Ember scores 179 and Gemini 3.8 Flash 173. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
179161–196
WeightsClosed
ReleasedNov 2026
Price in / out$0.10 / $0.50
Context1.05M
Core tests taken2 of 9
173157–189
WeightsClosed
ReleasedSep 2026
Price in / out$0.82 / $4.13
Context1.05M
Core tests taken7 of 9
Muse Ember and Gemini 3.8 Flash are too close to call (6 points apart, within the uncertainty), ahead on 1 of 2 shared core tests, and costs 8.2× less.
Test by testthird-party results
Every result as a table
| Test | Muse Ember | Gemini 3.8 Flash |
|---|---|---|
| Reasoning & knowledge | ||
| SimpleBench | 72.7% | 82.4% |
| Coding | ||
| Terminal-Bench 4.0 | 39.4% | 19.7% |
| Vibe Code Bench v1.1 | 83.7% | 78.6% |
| Agents | ||
| APEX-Agents | 64.3% | 64.3% |
| Professional work | ||
| HLAB-Vals-1.1 | 25.0% | 12.1% |
| Legal Research Bench | 49.0% | 38.9% |
| Tax Agent Bench | 38.2% | 32.1% |
Tests at least two of these models have taken. Bold rows are the core tests.