Compare / Ministral 3 8B 2512 vs Kimi K2 0905
Ministral 3 8B 2512 vs Kimi K2 0905
12 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Ministral 3 8B 2512 | Kimi K2 0905 |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 5.50 | 15.3 |
| Artificial Analysisreported | GPQA Diamond% of questions | 47.1% | 76.7% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 4.3% | 6.4% |
| Artificial Analysisreported | IFBench% of instructions | 29.1% | 41.7% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 25.7% | 53.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 4.5% | 23.5% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 26.6% | 73.4% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 23.0% | 22.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.33 | 0.31 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 28.8 | 54.4 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 99.1% | 98.6% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 21.7% | 17.9% |
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.