Compare / grok-4.20-0309-reasoning vs GLM 5.1
grok-4.20-0309-reasoning vs GLM 5.1
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 | grok-4.20-0309-reasoning | GLM 5.1 |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 25.2 | 26.1 |
| Artificial Analysisreported | GPQA Diamond% of questions | 88.5% | 86.8% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 32.4% | 30.1% |
| Artificial Analysisreported | IFBench% of instructions | 82.9% | 76.3% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 67.7% | 73.7% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 40.9% | 43.2% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 96.5% | 97.7% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 30.2% | 34.0% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 46.0% | 9.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.33 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 55.3 | 61.7 |
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.