Claude Fable 5 vs Claude Sonnet 5.5

$8.75 / task
Wins 2 of 21 benchmarks
$0.88 / task
Wins 6 of 21 benchmarks
Fable 554.1
Sonnet 5.559.4
Fable 51,609.9
Sonnet 5.51,558.1
Fable 548.8
Sonnet 5.5at max57.5
Fable 542.4
Sonnet 5.543.9
Fable 51,536.9
Sonnet 5.51,634.1
Fable 582.3
Sonnet 5.578
Fable 569.7
Sonnet 5.5at max70.5
Fable 535.8
Sonnet 5.5at max75
Fable 539.7
Sonnet 5.5—
Fable 538.1
Sonnet 5.5—
Fable 5—
Sonnet 5.534.3
Fable 551.1
Sonnet 5.5—
Fable 5—
Sonnet 5.544.3
Harvey LAB-AA 1.1?
Fable 5—
Sonnet 5.5at max2.8
Fable 5—
Sonnet 5.5at xhigh95.8
Fable 5—
Sonnet 5.5at xhigh70.6
EBR Bench?
Fable 539.5
Sonnet 5.5—
SpireClimb?
Fable 5—
Sonnet 5.5at max1.3
SpireClimb StS1 A20 Heart?
Fable 5—
Sonnet 5.5at max0
SpireClimb StS2 A10?
Fable 5—
Sonnet 5.5at max2.7

Specs

SpecFable 5Sonnet 5.5
Input, per 1M$10$2
Output, per 1M$50$10
Cost per task?$8.75$0.88
Speed, tok/s67106
Context1M1M
WeightsClosedClosed
ReleasedJun 2026Sep 2026
See images✓✓
Read PDFsPartial✓
Hear audio——
Watch video——
Call tools✓✓
Force a tool callPartial—
Return strict JSON✓✓
Search the web?✓
Run code?✓
Use a computer?✓

What breaks when you switch

Fable 5 → Sonnet 5.5
No forced tool use›

Forced tool_choice any or tool returns HTTP 400; remove the forced choice.

Replace disabled thinking›

Disabled thinking returns HTTP 400; use between_tools for the lowest thinking setting.

No manual thinking budgets›

Manual thinking budgets and extra between_tools fields return HTTP 400; remove manual budgets and send between_tools without display, budget_tokens or block_binding.

Remove sampling parameters›

Non-default temperature, top_p or top_k returns HTTP 400; remove these parameters.

No assistant prefill›

A prefilled final assistant turn returns HTTP 400; remove that prefill.

Old computer tool rejected · Claude API / Google Cloud›

The computer_20251124 tool returns HTTP 400 on the Claude API and Google Cloud; use computer_toolset_20260801 there.

Sonnet 5.5 → Fable 5
Nothing known.