AI Radar

llm-anthropic 0.31 adds Fable 5.2 and Mythos 5.2

First seen Nov 9 Updated Nov 10, 17:18 UTC Heat 45 medium evidence

Simon Willison released llm-anthropic 0.31, which adds Fable 5.2 and Mythos 5.2 to his llm command-line tool. He reports that Fable 5.2 at max reasoning effort ran out of tokens after 128,000 thinking tokens and returned no image in his pelican test. Fable 5 and 5.1 did not show this at max effort.

What to do
Run llm install llm-anthropic, set your Anthropic key, then run llm -m claude-fable-5.2.
Who should care
Developers who use the llm CLI or call Anthropic models through the API.

Why it matters: Fable 5.2 costs $8 per million input tokens and $40 per million output tokens, 20% below Fable 5.1, and it works in the llm CLI after installing llm-anthropic 0.31.

Key facts

llm-anthropic version
0.31
Model ID
claude-fable-5.2
Price
$8 input / $40 output per million tokens
Fable 5.1 price was $10 / $50
Max effort failed after 128,000 thinking tokens

Context

Willison says the max-effort failure matches a bug he saw with Opus 5.5.

Timeline

Coverage

  1. llm-anthropic 0.31simonwillison.net ยท Simonwillison

Evidence

Medium evidence: several community reports point the same way, but there is no official confirmation yet.

Details written Nov 10, 17:17 UTC from the sources listed above.