Claude Fable 5.2
- System card dated 9 November 2026
- Framework: Responsible Scaling Policy (RSP) and Frontier Compliance Framework (FCF)
- 12 tests by the provider
- 1 source document
The provider's own tests
- Bias Benchmark for Question Answering (BBQ) · The provider · Bias
On disambiguated questions, Fable 5.2 answered 89.69% of questions correctly, in line with Claude Fable 5.1 (89.92%) and Claude Opus 5.5 (89.65%).
Anthropic · 9 November 2026 · Read the source - Indirect Prompt Injection (IPI) benchmark · The provider · Injection & agentic
Reported result: Attack success rate of 0.30% at k=1, 2.90% at k=10, and 4.09% at k=15; at k=15, 0.68% in coding, 4.37% in tool use, and 9.18% in GUI computer use.
Claude Fable 5.2 has an attack success rate of 0.30% at k=1, 2.90% at k=10, and 4.09% at k=15.
Anthropic · 9 November 2026 · Read the source - LMCA · The provider · Autonomy & replication
[Claude Mythos 5.2] represents a small improvement above Fable 5.1 and Claude Opus 5.5 on LMCA, but not a clear trend break.
Anthropic · 9 November 2026 · Read the source - Shade · The provider · Injection & agentic
Reported result: 89.82% attack success rate without safeguards and 28.16% with probes enabled, with 40/40 scenarios compromised in both cases.
Without prompt injection probes, the attacker has a success rate of 89.82% on Claude Fable 5.2, higher than on Claude Opus 5.5 (54.61%) and Claude Fable 5.1 (51.93%).
Anthropic · 9 November 2026 · Read the source - agentic safety evaluations · The provider · Injection & agentic
Our agentic safety evaluations found that Claude Fable 5.2 falls back to Opus 4.8 more frequently than recent models do in prompt injection scenarios, resulting in a higher attack success rate.
Anthropic · 9 November 2026 · Read the source - multi-turn election integrity evaluation · The provider · Bias
On multi-turn adversarial test cases, Fable 5.2’s appropriate response rate was slightly below that of Claude Fable 5.1 on the API (85% vs. 90%) and slightly above it on claude.ai (90% vs. 88%).
Anthropic · 9 November 2026 · Read the source - multi-turn evaluations (disordered eating) · The provider · Harmful content
In a review of the multi-turn evaluations, internal policy experts found that Fable 5.2 was less likely than Fable 5.1 to give unfocused or overwhelming replies, to make claims about the user that the user had not supported, to assign a definitive diagnosis, and to suggest well-meant behavioral changes that may be harmful in the context of disordered eating.
Anthropic · 9 November 2026 · Read the source - open-source political even-handedness evaluation · The provider · Bias
Fable 5.2 responded even-handedly on nearly all 1,350 pairs of prompts representing opposing political perspectives, and scored slightly higher than Claude Fable 5.1 (99.1% vs. 98.3% on the API, and 99.7% vs. 98.8% on claude.ai ), though the difference on the API was within the margin of error.
Anthropic · 9 November 2026 · Read the source - replayed coding sessions · The provider · Autonomy & replication
In replayed coding sessions, it left doable, requested work undone in 6–13% of sessions at medium to xhigh effort, compared with 4% for Claude Fable 5.1.
Anthropic · 9 November 2026 · Read the source - single-turn election integrity benchmark · The provider · Bias
On the single-turn evaluation, Fable 5.2 declined nearly every violative prompt and complied with all of the harmless prompts, both on the core model (without a system prompt) and on claude.ai , in line with the other models we evaluated.
Anthropic · 9 November 2026 · Read the source
The 10 newest are listed above; show all 12
- single-turn harmful and benign request evaluation for disordered eating · The provider · Harmful content
On single-turn evaluations, Claude Fable 5.2 performed comparably to Claude Fable 5.1 on requests that could reinforce disordered eating, both on the API (94.48% vs. 95.69%, within the margin of error) and on claude.ai (99.50% vs. 99.43%), and it did not overrefuse benign requests about nutrition, fitness, or body image on either surface.
Anthropic · 9 November 2026 · Read the source - suicide and self-harm evaluations · The provider · Harmful content
In suicide and self-harm conversations, Fable 5.2 was less likely to suggest harmful substitution behaviors, but it still showed some areas for improvement on the API, most of which were mitigated by the claude.ai system prompt.
Anthropic · 9 November 2026 · Read the source
No document we read states: a release date, whether a capability level was crossed, an independent or government test, a safeguard. No incident on record names this version.
Documents listed for this version
- System Card: Claude Fable 5.2 & Claude Mythos 5.2 · Anthropic · 9 November 2026