Compare / GPT-5 Mini vs grok-4-1-fast-non-reasoning
GPT-5 Mini vs grok-4-1-fast-non-reasoning
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 | GPT-5 Mini | grok-4-1-fast-non-reasoning |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 20.6 | 11.3 |
| Artificial Analysisreported | GPQA Diamond% of questions | 82.8% | 63.7% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 21.5% | 5.1% |
| Artificial Analysisreported | IFBench% of instructions | 75.4% | 36.5% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 72.3% | 31.3% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 33.3% | 14.4% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 71.1% | 63.7% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 91.0% | 74.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 44.3% | 26.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 27.5% | 46.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 55.5% | 58.3% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 93.8% | 81.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 69.8% | 88.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 62.5% | 77.9% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 82.0% | 75.0% |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 99.9% | 98.5% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 12.9% | 17.8% |
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.