Claude Mythos 5.1
18 values about Claude Mythos 5.1 from 2 documents, in 7 metric families. In each family, values from its own document come first.
- Developer
- Anthropic
- Release date
- 1 Sep 2026The date of its first card in the dataset, the Claude Fable 5.1 & Claude Mythos 5.1 System Card.
- Availability
- Public
- Its own document
- Claude Fable 5.1 & Claude Mythos 5.1 System Card · 1 Sep 2026
- Also reported in
- Claude Opus 5.5 System Card · 22 Sep 2026
Data as of 26 Sep 2026 · Dataset v0.1 · Methodology v0.1 · Changelog
Values by metric family
Each value as the document printed it. Select a value for its source, its checks and its history. Values from other documents are listed after the model’s own, each marked as the first report of that measure or as a restatement.
KeyVerifiedUnverifiedDisputedCorrected read off a figure or stated in wordsA value opens its source and history.
M2 · Reward hacking 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentClaude Fable 5.1 & Claude Mythos 5.1 System Card · 1 Sep 2026 | ||
| Reward hacking in RL trainingShare of episodes with successful reward hacks | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
M3 · Sabotage and sandbagging 3 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentClaude Fable 5.1 & Claude Mythos 5.1 System Card · 1 Sep 2026 | ||
| LinuxArenaStealth success vs Opus 4.8 monitorCondition: extended thinking | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| LinuxArenaStealth success vs Opus 4.8 monitorCondition: no extended thinking | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| SHADE-ArenaStealth success vs Mythos Preview | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
M4 · Misalignment audits 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentClaude Fable 5.1 & Claude Mythos 5.1 System Card · 1 Sep 2026 | ||
| Internal deployment monitoringClassifier workarounds / disabled checks | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
M5 · Honesty and hallucination 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentClaude Fable 5.1 & Claude Mythos 5.1 System Card · 1 Sep 2026 | ||
| MASKHonesty rate under pressure | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
M10 · Dangerous capabilities and risk determinations 10 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentClaude Fable 5.1 & Claude Mythos 5.1 System Card · 1 Sep 2026 | ||
| Anthropic ECI (AECI)Capability index | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| DNA synthesis screening evasionPathogens meeting all criteriaCondition: of 10 pathogens | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| ExploitBenchMean flags capturedCondition: AutoNudge arm | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| Firefox 147 exploitationFull exploit rate | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| Long-form virology task 1Score | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| Long-form virology task 2Score | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| Multimodal virology (VCT)Score | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| OSS-FuzzTargets with full scoreCondition: ~830 entry points | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
| From other documents | ||
| Anthropic ECI (AECI)Capability indexClaude Opus 5.5 System Card22 Sep 2026First reported | Claude Opus 5.5 System Card22 Sep 2026First reported | |
| CoBenchScoreClaude Opus 5.5 System Card22 Sep 2026First reported | Claude Opus 5.5 System Card22 Sep 2026First reported | |
M11 · Self-preservation 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentClaude Fable 5.1 & Claude Mythos 5.1 System Card · 1 Sep 2026 | ||
| Sandbox vulnerability exploitReal-world incident reported | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |
M12 · Chain-of-thought monitorability 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentClaude Fable 5.1 & Claude Mythos 5.1 System Card · 1 Sep 2026 | ||
| CoT controllabilityChange vs Mythos PreviewCondition: best-performing prompt | Claude Fable 5.1 & Claude Mythos 5.1 System Card1 Sep 2026 | |