Modellvergleich
Wählen Sie bis zu 4 Modelle aus und vergleichen Sie sie über alle Metriken.
Beliebte Duelle:
| Metrik | Claude Mythos Preview | Claude Fable 5 | Claude Fable 5.1 (XHigh) |
|---|---|---|---|
| Anbieter | Anthropic | Anthropic | Anthropic |
| Lizenz | Proprietary | Proprietary | Proprietary |
| Podium Score | 97.4 | 93.4 | 89.0 |
| Gesamtrang | 1 | 2 | — |
| Arena Elo | — | 1509 | — |
| Intelligenz-Index | — | 60.0 | — |
| Coding | 80.9 | 84.1 | 88.0 |
| Mathematik | 100.0 | — | — |
| Reasoning | 98.8 | 81.0 | 85.0 |
| Agentic | — | 96.2 | — |
| Wissen | 100.0 | 43.3 | 45.0 |
| Multimodal | 100.0 | — | 30.0 |
| Langkontext | — | 99.2 | — |
| Tempo | — | 10.2 | 40.0 |
| Preis-Leistung | — | 71.4 | 40.0 |
| Ausgabegeschwindigkeit | 80 t/s | 71 t/s | 78 t/s |
| Zeit bis zum ersten Token | — | 141 ms | 350 ms |
| Eingabepreis ($/M) | $3/M | $10/M | $3/M |
| Ausgabepreis ($/M) | $15/M | $50/M | $15/M |
| Kontextfenster | 1000K | 1000K | 500K |
| SWE-Bench Verified | 93.9% | 95% | — |
| GPQA Diamond | 94.6% | 92.6% | — |
| Multilingual MMLU | 92.7% | — | — |
| Browsecomp | 86.9% | — | — |
| CharXiv Reasoning | 93.2% | — | — |
| Humanity’s Last Exam | 64.7% | 53.3% | — |
| SWE-Bench Pro | 77.8% | — | — |
| SciCode | — | 60.2% | — |
| LiveCodeBench | — | 70% | — |
| Ifbench | — | 63.5% | — |
| τ²-Bench Retail | — | 98.5% | — |
| Terminal-Bench Hard | — | 62.9% | — |