Claude Fable 5.1 & Claude Mythos 5.1 System Card
A system card by Anthropic about Claude Fable 5.1 and Claude Mythos 5.1, published 1 Sep 2026. We recorded 22 values from it.
- Developer
- Anthropic
- Type
- System card
- Models covered
- Claude Fable 5.1, Claude Mythos 5.1
- Published
- 1 Sep 2026
- Archived copy
- No archived copy yet
- Changelog
- Not known
Data as of 26 Sep 2026 · Dataset v0.1 · Methodology v0.1 · Changelog
Versions
One version is on record: the copy we retrieved on 26 Sep 2026. We know of no other.
26 Sep 2026
Date retrieved
Copy retrieved 26 Sep 2026
No version has a file hash or an archived snapshot yet. From dataset v0.2 each retrieved version carries both (Methodology §7).
Revisions
No revisions are recorded for this document. We know of only one version of it.
Values
Every value we recorded from this document, grouped by metric family and ordered by where the document prints it. Location is the section, table or page as the document numbers it. None has been blind-verified yet.
KeyVerifiedUnverifiedDisputedCorrected read off a figure or stated in wordsA value opens its source and history.
M2 Reward hacking
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5.1 | Reward hacking in RL trainingShare of episodes with successful reward hacks | None stated | Sec 6 | Unverified |
M3 Sabotage and sandbagging
3 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5.1 | LinuxArenaStealth success vs Opus 4.8 monitor | extended thinking | Sec 2.4.1 | Unverified | |
| Claude Mythos 5.1 | LinuxArenaStealth success vs Opus 4.8 monitor | no extended thinking | Sec 2.4.1 | Unverified | |
| Claude Mythos 5.1 | SHADE-ArenaStealth success vs Mythos Preview | None stated | Sec 2.4.1 | Unverified |
M4 Misalignment audits
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5.1 | Internal deployment monitoringClassifier workarounds / disabled checks | None stated | Sec 2.4 / 6.1 | Unverified |
M5 Honesty and hallucination
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5.1 | MASKHonesty rate under pressure | None stated | Sec 6 | Unverified |
M7 Harmful compliance and over-refusal
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Fable 5.1 | Multi-turn biological weaponsSafe response rate | API | Sec 4.1.3 | Unverified |
M9 Prompt injection
2 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Fable 5.1 | Browser use prompt injectionAttack success rate | None stated | Sec 5.2 | Unverified | |
| Claude Fable 5.1 | Computer-use prompt injectionAttack success rate | None stated | Sec 5.2 | Unverified |
M10 Dangerous capabilities and risk determinations
11 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Fable 5.1 | RSP determinationsCB / autonomy / cyber | None stated | Sec 1-2 | Unverified | |
| Claude Mythos 5.1 | DNA synthesis screening evasionPathogens meeting all criteria | of 10 pathogens | Sec 2.2.3.1 | Unverified | |
| Claude Mythos 5.1 | Long-form virology task 1Score | None stated | Sec 2.2.3.1 | Unverified | |
| Claude Mythos 5.1 | Long-form virology task 2Score | None stated | Sec 2.2.3.1 | Unverified | |
| Claude Mythos 5.1 | Multimodal virology (VCT)Score | None stated | Sec 2.2.3.1 | Unverified | |
| Claude Mythos 5 | Anthropic ECI (AECI)Capability index | None stated | Sec 2.3.5 | Unverified | |
| Claude Mythos 5.1 | Anthropic ECI (AECI)Capability index | None stated | Sec 2.3.5 | Unverified | |
| Claude Opus 5 | Anthropic ECI (AECI)Capability index | None stated | Sec 2.3.5 | Unverified | |
| Claude Mythos 5.1 | ExploitBenchMean flags captured | AutoNudge arm | Sec 3.3.1 | Unverified | |
| Claude Mythos 5.1 | OSS-FuzzTargets with full score | ~830 entry points | Sec 3.3.2 | Unverified | |
| Claude Mythos 5.1 | Firefox 147 exploitationFull exploit rate | None stated | Sec 3.3.3 | Unverified |
M11 Self-preservation
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5.1 | Sandbox vulnerability exploitReal-world incident reported | None stated | Sec 2.4 / 6 | Unverified |
M12 Chain-of-thought monitorability
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5.1 | CoT controllabilityChange vs Mythos Preview | best-performing prompt | Sec 2.4.1 | Unverified |
Extraction coverage
What we read of this document, and where each value was read.
Our note Web reader stopped ~p.52; refers to an August 2026 Risk Report (not extracted)
Of the 22 values, 17 were read in the document itself, 1 on a developer summary page that restates it and 4 in an independent write-up that quotes it.
Values read somewhere other than the document stand in where the document's own section could not be read directly, and are flagged on every value (Methodology §2.2).
Developer summary pages used
Independent write-ups used
Of the 22 values, 2 are statements in words rather than numbers; they are marked * and left out of charts by default.
All 22 values were extracted for version 0 of the dataset through a web reader, which did not always reach the later sections of long PDFs (Methodology §3).