Claude Fable 5
- Made by
- Anthropic
- Released
- Announced in
- Claude Fable 5 and Claude Mythos 5anthropic.com
- Event
- Claude Fable 5 and Mythos 5
- API record
- /api/v1/models/claude-fable-5
Benchmark results
| Benchmark | Score | Source check |
|---|---|---|
| Humanity's Last ExamNo tools · reported later, in Introducing Claude Opus 5 | 56.5% | CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.CorroboratedThe same number is printed in 1 other lab document: 1 |
| reported later, in Introducing Claude Fable 5.1 and Claude Mythos 5.1 | 57.8% | CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Humanity's Last ExamWith tools · reported later, in Introducing Claude Opus 5 | 63.9% | CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.CorroboratedThe same number is printed in 1 other lab document: 1 |
| reported later, in Introducing Claude Fable 5.1 and Claude Mythos 5.1 | 63.8% | CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| SWE-bench Verified | 95% | CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.CorroboratedThe same number is printed in 1 other lab document: 1 |
| SWE-bench Proadaptive thinking, max effort; average of 5 trials; with production safeguards, including fallback to Opus 4.8 · The source doesn't say whether this is the public or the commercial set | 80% | CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| BrowseCompreported later, in Introducing Claude Opus 5 | 87.4% | CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| AutomationBenchZapier leaderboard, private held-out set; max effort | 17.4% | |
| Zapier leaderboard, private held-out set; tasks blocked by safeguards scored zero · reported later, in Introducing Claude Fable 5.1 and Claude Mythos 5.1 | 17.1% | CorroboratedThe same number is printed in 1 other lab document: 1 |
| DeepSWE v1.1reported later, in Introducing Claude Opus 5 | 69.7% | CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
Sources
Reported by Anthropic
- Claude Fable 5 and Claude Mythos 5 System Cardanthropic.com · 9 Jun 2026
Our capture of page 251 
Our capture of pages 251-252 - Introducing Claude Opus 5anthropic.com · 24 Jul 2026
Original image - Introducing Claude Fable 5.1 and Claude Mythos 5.1anthropic.com · 1 Sep 2026
Our capture of the source
