Compare / Grok 4.20 vs GLM 5.2
Grok 4.20 vs GLM 5.2
10 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 | GLM 5.2 |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 25.7 | 33.7 |
| Artificial Analysisreported | GPQA Diamond% of questions | 91.1% | 89.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 34.5% | 41.1% |
| Artificial Analysisreported | IFBench% of instructions | 81.2% | 73.3% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 69.0% | 78.3% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 37.9% | 50.8% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 93.0% | 99.1% |
| tau2-benchreported | Consistency (pass^4)% of tasks | 8.2% | 13.4% |
| tau2-benchreported | Task success (pass^1)% of tasks | 18.0% | 37.1% |
| Vending-Bench 2reported | Money after a yearUS dollars | $4,662.85 | $8,313.78 |
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.