Claude Fable 5 & Claude Mythos 5 System Card
A system card by Anthropic about Claude Fable 5 and Claude Mythos 5, published 9 Jun 2026. We recorded 20 values from it.
- Developer
- Anthropic
- Type
- System card
- Models covered
- Claude Fable 5, Claude Mythos 5
- Published
- 9 Jun 2026
- Archived copy
- No archived copy yet
- Changelog
- Has a changelog
Data as of 26 Sep 2026 · Dataset v0.1 · Methodology v0.1 · Changelog
Revisions
Each change we have recorded between two versions: the old and the new text, whether the document's changelog explains it, how we know, and whether we have checked it. A revision is not in itself evidence of wrongdoing; most are corrections (Methodology §7).
lowvery lowExplained — corrected to match §2.4
How we know Changelog read 26 Sep 2026
Checked Confirmed from changelog
No copy of the 11 Jun 2026 version held No copy of the 25 Jun 2026 version held
Versions
Four versions are on record, oldest first. We hold a copy of one; the others are known only by their date.
9 Jun 2026
First published version
11 Jun 2026
Revision
What changed several corrections
25 Jun 2026
Revision
What changed 1 change recorded as a revision. See it in redline
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).
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. 2 of the 20 have been blind-verified: a second reader found the same value without seeing ours.
KeyVerifiedUnverifiedDisputedCorrected read off a figure or stated in wordsA value opens its source and history.
M1 Evaluation awareness
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5 | Grader awarenessShare of episodes aware of grader | grader at high risk of hacking | Sec 6 | Unverified |
M4 Misalignment audits
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5 | Internal deployment monitoringMisaligned actions in monitored traffic | None stated | Sec 6 | Unverified |
M5 Honesty and hallucination
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5 | Input hallucinationRate of hallucinating missing reference | None stated | Sec 6.3.3.4 | Unverified |
M8 Jailbreak robustness
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Fable 5 | Cyber safeguards automated red-teamingTask completion by automated red-teamer | with Fable safeguards | Sec 4/5 | Unverified |
M9 Prompt injection
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos 5 | External prompt injection benchmarkRank vs prior models | None stated | Exec summary | Unverified |
M10 Dangerous capabilities and risk determinations
15 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Fable 5 | Release decisionDeployment scope | None stated | Sec 1 | Unverified | |
| Claude Mythos 5 | Autonomy threat model 2 (automated R&D)RSP determination | None stated | Sec 2 | Unverified | |
| Claude Mythos 5 | CB-1 threat modelRSP determination | None stated | Sec 2 | Unverified | |
| Claude Mythos 5 | CB-2 threat modelRSP determination | None stated | Sec 2 | Unverified | |
| Claude Mythos 5 | DNA synthesis screening evasionPathogens with viable evasion | of 10 pathogens | Sec 2.2 | Unverified | |
| Claude Mythos 5 | Long-form virology task 1Score | None stated | Sec 2.2 | Unverified | |
| Claude Mythos 5 | Long-form virology task 2Score | None stated | Sec 2.2 | Unverified | |
| Claude Mythos 5 | Multimodal virology (VCT)Score | None stated | Sec 2.2 | Unverified | |
| Claude Mythos 5 | Anthropic ECI (AECI)Capability index | None stated | Table 2.3.5 | Unverified | |
| Claude Mythos Preview | Anthropic ECI (AECI)Capability index | None stated | Table 2.3.5 | Unverified | |
| Claude Mythos 5 | AI R&D LLM training taskAverage speedup | None stated | Table 2.3.7.1.A | Verified | |
| Claude Mythos Preview | AI R&D LLM training taskAverage speedup | re-run | Table 2.3.7.1.A | Verified | |
| Claude Opus 4.8 | AI R&D LLM training taskAverage speedup | re-run | Table 2.3.7.1.A | Unverified | |
| Claude Mythos 5 | AI R&D kernel taskBest speedup | None stated | Table 2.3.7.A | Unverified | |
| Claude Mythos 5 | AI R&D novel compiler taskPass rate | None stated | Table 2.3.7.A | Unverified |
Extraction coverage
What we read of this document, and where each value was read.
Our note Web reader stopped ~p.50
Of the 20 values, 16 were read in the document itself, 1 on a developer summary page that restates it and 3 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 20 values, 1 is a statement in words rather than a number; it is marked * and left out of charts by default.
All 20 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).