Claude Fable 5.1 made me a really nice animated pelican
- ID
- 20486
- Status
- summarized
- Published
- 02 Sep 2026, 7:57 AM
- Fetched
- 02 Sep 2026, 8:30 AM
- Provider
- Simon Willison
- Category
- developer-ai
- Original URL
- https://simonwillison.net/2026/Sep/1/claude-fable-5-1/
- Source URL
- https://simonwillison.net/atom/everything/
Summary
- Score
- 7.0
- Created
- 02 Sep 2026, 8:30 AM
- Tags
- Audience
- developersvibe_codersai_agent_users
What happened
Simon Willison tested Anthropic's newly released Claude Fable 5.1 using his informal 'pelican benchmark' (generating an SVG of a pelican on a bicycle) across all five reasoning effort levels: low, medium, high, xhigh, and max. Notably, at low and medium effort, Fable 5.1 appeared to skip reasoning entirely for this prompt, producing similar output at ~10 cents and ~24 seconds. At xhigh, costs exploded to $1.83 for a single SVG, taking 7 minutes 51 seconds and consuming 36,767 output tokens. Anthropic's headline claim is a 52.6% score on the brand-new Terminal-Bench-Science 0.1 benchmark, up from 24.7% for Fable 5.
Why it matters
If you use Claude for coding or generation tasks, reasoning effort level now has dramatic cost and latency implications: jumping from medium to xhigh turned a 10-cent, 24-second task into a $1.83, 8-minute task. You should test which reasoning level you actually need for your workload before defaulting to high or above, especially for batch or agentic pipelines where costs compound. Also note there is no longer an option to fully disable reasoning.
Discussion angle
The reasoning-level cost cliff is the real story: for a simple SVG, xhigh cost 18x more than medium with unclear quality difference. What's the right default reasoning level for agentic coding workflows, and how do you detect when a task actually warrants xhigh or max?