Claude Fable 5.2
Claude Fable 5.2 is an API-only (closed weights) language model from Anthropic, released November 9, 2026. The context window is 1M tokens. On the Epoch Capabilities Index it scores 173.4, rank 1 of 256 rated models. API price: $8 input / $40 output per million tokens.
Check it against your own PC →
Lab
Anthropic
Released
November 9, 2026
Weights
API-only (closed weights)
Parameters
–
Context
1M
Price / M tokens
$8 / $40 in / out
Epoch ECI
173.4 #1
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Reasoning | 62.1 | Chess Puzzles, SimpleBench, Mystery Game Puzzles |
| Math | 81.5 | ProofBench, OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private |
| Knowledge | 88.0 | GPQA diamond, SimpleQA Verified |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| ProofBench | 100.0% |
| FrontierMath (Tiers 1–3) | 100.0% |
| FrontierMath-Tier-4-v2-Private | 100.0% |
| OTIS Mock AIME | 99.7% |
| GPQA Diamond | 95.7% |
| SimpleQA Verified | 94.0% |
| SimpleBench | 93.9% |
| EBR-bench | 90.5% |
| Mystery Game Puzzles | 89.0% |
| Furniture Assembly | 85.7% |
| Chess Puzzles | 77.9% |
| FrontierSWE | 69.9% |
| Remote Labor Index | 41.3% |
| SpireClimb | 19.6% |
Links
Models with a similar score
More from Anthropic
Claude Haiku 5.5 · Claude Sonnet 5.5 · Claude Opus 5.5 · Claude Fable 5.1 · Claude Opus 5 · Claude Sonnet 5 · Claude Fable 5 · Claude Mythos 5