Compare / Mistral Small 4 vs GPT-6 Astra
Mistral Small 4 vs GPT-6 Astra
17 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Mistral Small 4 | GPT-6 Astra |
|---|---|---|---|
| BulletBenchmeasured by us | Cost per gameUS dollars | $0.0039 | $0.0943 |
| BulletBenchmeasured by us | Games lost on time% of games | 25.0% | 75.0% |
| BulletBenchmeasured by us | Invalid moves% of moves | 0.6% | 0.0% |
| BulletBenchmeasured by us | Ladder Eloladder Elo | 367 | 323 |
| BulletBenchmeasured by us | Median move timemilliseconds | 0.6 s | 2.3 s |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 26.6 | 77.1 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 11.3 | 52.7 |
| Artificial Analysisreported | GPQA Diamond% of questions | 76.9% | 96.3% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 9.9% | 54.7% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 49.7% | 80.7% |
| Artificial Analysisreported | SciCode% of problems | 38.8% | 56.5% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 21.0% | 89.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 0.0% | 59.6% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 4.9% | 43.1% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 22.0% | 0.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.31 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 40.3 | 68.9 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.