Compare / Claude Fable 5 vs Grok 4.20
Claude Fable 5 vs Grok 4.20
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 | Claude Fable 5 | Grok 4.20 |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 49.6 | 25.7 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.6% | 91.1% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 55.5% | 34.5% |
| Artificial Analysisreported | IFBench% of instructions | 63.5% | 81.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 82.3% | 69.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 62.9% | 37.9% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 98.5% | 93.0% |
| tau2-benchreported | Consistency (pass^4)% of tasks | 28.9% | 8.2% |
| tau2-benchreported | Task success (pass^1)% of tasks | 39.7% | 18.0% |
| Vending-Bench 2reported | Money after a yearUS dollars | $5,680.26 | $4,662.85 |
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.