Claude Opus 4.7 System Card
A system card by Anthropic about Claude Opus 4.7, published 16 Apr 2026. We recorded 24 values from it.
- Developer
- Anthropic
- Type
- System card
- Model covered
- Claude Opus 4.7
- Published
- 16 Apr 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. 3 of the 24 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
2 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Opus 4.6 | Automated behavioral auditShare of transcripts verbalizing eval awareness | None stated | Sec 6 | Unverified | |
| Claude Opus 4.7 | Automated behavioral auditShare of transcripts verbalizing eval awareness | None stated | Sec 6 | Verified |
M2 Reward hacking
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Opus 4.7 | Impossible tasksReward hack rate | None stated | Sec 6 | Unverified |
M3 Sabotage and sandbagging
4 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Mythos Preview | SHADE-ArenaStealth success rate | extended thinking | Sec 2.4.1 | Unverified | |
| Claude Opus 4.6 | SHADE-ArenaStealth success rate | extended thinking | Sec 2.4.1 | Unverified | |
| Claude Opus 4.7 | Minimal-LinuxBenchStealth success rate | CoT visible to monitor | Sec 2.4.1 | Unverified | |
| Claude Opus 4.7 | SHADE-ArenaStealth success rate | extended thinking | Sec 2.4.1 | Unverified |
M7 Harmful compliance and over-refusal
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Opus 4.7 | Single-turn benign requestsOver-refusal rate | None stated | Sec 4.1.2 | Unverified |
M10 Dangerous capabilities and risk determinations
15 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Opus 4.7 | CB-1 threat modelRSP determination | None stated | Sec 2 | Unverified | |
| Claude Opus 4.7 | CB-2 threat modelRSP determination | None stated | Sec 2 | Unverified | |
| Claude Opus 4.7 | DNA synthesis screening evasionPathogens with viable evasion | of 10 pathogens | Sec 2.2.5.2 | Verified | |
| Claude Opus 4.7 | Long-form virology task 1Score | None stated | Sec 2.2.5.2 | Unverified | |
| Claude Opus 4.7 | Long-form virology task 2Score | None stated | Sec 2.2.5.2 | Unverified | |
| Claude Opus 4.7 | Multimodal virology (VCT)Score | None stated | Sec 2.2.5.2 | Unverified | |
| Claude Opus 4.7 | Autonomy threat model 2 (automated R&D)RSP determination | None stated | Sec 2.3 | Unverified | |
| Claude Mythos Preview | AI R&D LLM training taskAverage speedup | None stated | Sec 2.3.4 | Verified | |
| Claude Opus 4.7 | AI R&D kernel taskBest speedup | None stated | Sec 2.3.4 | Unverified | |
| Claude Opus 4.7 | AI R&D LLM training taskAverage speedup | None stated | Sec 2.3.4 | Unverified | |
| Claude Opus 4.7 | AI R&D novel compiler taskPass rate | None stated | Sec 2.3.4 | Unverified | |
| Claude Opus 4.7 | AI R&D quadruped RL taskHighest score | None stated | Sec 2.3.4 | Unverified | |
| Claude Opus 4.7 | CybenchSuccess rate, pass@1 | 35 challenges | Sec 3.3.1 | Unverified | |
| Claude Opus 4.7 | CyberGymTargeted vuln reproduction rate | None stated | Sec 3.3.2 | Unverified | |
| Claude Mythos Preview | Cyber rangeFull range solvesRun by UK AI Security Institute | 10 attempts | Sec 3.4 | Unverified |
M12 Chain-of-thought monitorability
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Claude Opus 4.7 | Accidental CoT supervision in RLShare of training episodes affected | None stated | Sec 2.4 / 6 | Unverified |
Extraction coverage
What we read of this document, and where each value was read.
Our note Web reader stopped ~p.52
Of the 24 values, 19 were read in the document itself, 2 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
All 24 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).