Chaptered summary is still being generated for this document. Showing a heuristic brief in the meantime.
Summary
63,024-word document condensed to 242 words. Anthropic · May 15, 2026
TL;DR
“This system card describes Claude Opus 4.7, a large language model from Anthropic. Overall, the model shows superior capabilities to those of its predecessor, Claude Opus 4.6, but weaker capabilities than those of our most powerful model, Claude Mythos Preview.”
Capability claim
- “We are releasing Opus 4.7 with a new set of cybersecurity safeguards.”
Safety findings
- “We believe our risk mitigations are sufficient to make catastrophic risk from non-novel chemical/biological weapons production very low but not negligible.”
- “We believe that catastrophic risk from novel chemical/biological weapons remains low (with substantial uncertainty). The overall picture is similar to the one from our most recent Risk Report.”
- “We believe that the overall risk is very low, and that this model in particular adds little to the risk picture we previously laid out for Claude Mythos Preview .”
Deployment scope
- “accessible to experts, we interpret a model’s performance on this task primarily based on the expert’s assessment of uplift.”
Limitations the lab flags
- “limitations included sycophantic agreement under pushback, verbose responses that buried actionable content, degraded reference accuracy, and overconfidence in the feasibility of synthesis steps.”
- “open questions — particularly around fully explaining the evaluation- awareness results — that they would have preferred more time to resolve; and that the internal-usage evidence base for this model was thinner than for some prior releases.”
Every italicized passage is a verbatim substring of the source document (checked deterministically after extraction). Field selection is heuristic — some quotes may lack surrounding context and some claims may be absent if no matching pattern appeared. For citation, open the source: original model card · source SHA f055e7ef9acc · version dated May 15, 2026.
Extracted Evaluations(48 results)
Sort by:0/48 rows fully reproducible (0%)
| Benchmark | Category | State | Score | Setup | Source |
|---|---|---|---|---|---|
| agent | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| agent | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| agent | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| agent | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
/ verified | coding | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
/ pro | coding | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
/ multilingual | coding | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
/ multimodal | coding | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
| multilingual | mentioned | — | Averageinstruction-tunedmissing: shot countmissing: method | self-reported | |
Firefox 147 | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
EARL | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
SHADE-Arena | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
Minimal-LinuxBench | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
Terminal-Bench | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
USAMO/ 2026 | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
MRCR/ v2 | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
Lab-Bench/ figqa | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
CharXiv/ reasoning | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
ScreenSpot/ pro | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
VendingBench | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
/ aa | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
GMMLU | other | mentioned | — | Averageinstruction-tunedmissing: shot countmissing: method | self-reported |
MILU | other | mentioned | — | Averageinstruction-tunedmissing: shot countmissing: method | self-reported |
INCLUDE | other | mentioned | — | Averageinstruction-tunedmissing: shot countmissing: method | self-reported |
BioPipelineBench/ verified | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
BioMysteryBench/ verified | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
Long-form virology tasks | other | mentioned | — | without-safeguardsinstruction-tunedmissing: shot countmissing: language | self-reported |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
Sequence-to-function/ modeling | other | mentioned | — | extended-thinkinginstruction-tunedmissing: shot countmissing: language | self-reported |
Sequence-to-function/ design | other | mentioned | — | extended-thinkinginstruction-tunedmissing: shot countmissing: language | self-reported |
AECI | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
Petri | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
Reward hacking evaluations | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
Automated Behavioral Audit | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
Claude self-preference evaluation | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
Decision theory evaluation | other | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
/ diamond | reasoning | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |
| reasoning | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported | |
| safety | mentioned | — | instruction-tunedmissing: shot countmissing: methodmissing: language | self-reported |