Nov 9, 2026
Anthropic's Claude Fable 5.2 is #1 on the Vals Index
We evaluated Anthropic’s new Claude Fable 5.2 on Vals Index v2.2, its components and the Vals Science Index.
-
#1 of 36 on the Vals Index (71.31%) at $17.94 per test, ahead of Claude Sonnet 5.5 (70.85%) and Gemini 4 Carbon (70.45%).
-
#1 on four Vals Index components: Finance Agent v2.1 (69.47%), Vibe Code Bench (93.77%), EMB v1.1 (77.38%) and, tied with Claude Opus 5.5, Terminal-Bench 4.0 (65.15%).
-
Also #1 of 66 on ProgramBench (19.00%, ahead of Opus 5.5 at 18.50%), and #2 of 22 on SWE-bench Multimodal (62.29%), behind Opus 5.5 (70.21%).
-
Strong on science: #1 of 31 on BioMysteryBench (83.70%), and #2 of 24 on the Vals Science Index (63.74%), behind GPT-6 Astra (65.09%).
-
Weaker on HLAB-Vals-1.1 (5.00%, tied #29 of 37) and Tax Agent Bench (32.09%, #27 of 84).
Claude Fable 5.2 is a reasoning model with a 1M-token context window, priced at $8.00 per million input tokens and $40.00 per million output tokens. We evaluated it at High effort.
Congrats to the Anthropic team on the release!