/AI4h ago

Claude Fable 5 Tops APEX-SWE Benchmark at 65.5% Pass@1

11275163424K
Original post
Nathan Lambert@natolambert#64inAI

A crazy jump. The price of the tokens will be worth it to a vast number of enterprises.

10:56 AM · Jun 9, 2026 · 11.1K Views
Sentiment

Users are excited about Claude Fable 5 topping the APEX-SWE benchmark because the 20pp gain over Opus delivers substantial value for real enterprise software engineering tasks.

Pos
100.0%
Neg
0.0%
1 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS31LIKES1
haro@harobuilds

@natolambert 20pp over opus 4.8 is not a marginal improvement. enterprises will pay whatever anthropic asks for that gap on real swe tasks

3hViews 31Likes 1
china232332@gigantictur

@natolambert https://x.com/mercor_ai/status/2064399136007589994?s=20 Is it just trained for tool cool better , but the way it definitley reviews code is very AGI/neuralese pilled

3hViews 25