Compare / GPT-5.4 vs grok-4.20-0309-reasoning
GPT-5.4 vs grok-4.20-0309-reasoning
11 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-5.4 | grok-4.20-0309-reasoning |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 39.0 | 25.2 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.0% | 88.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 43.7% | 32.4% |
| Artificial Analysisreported | IFBench% of instructions | 73.9% | 82.9% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 82.0% | 67.7% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 57.6% | 40.9% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 87.1% | 96.5% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 45.1% | 30.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 30.0% | 46.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.40 | 0.33 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 68.8 | 55.3 |
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.