FrontierThe story, in brief
Anthropic claims its new Claude Opus 5 delivers near-Fable 5 performance at half the token price
30.2% on ARC-AGI-3. Anthropic's Claude Opus 5 just matched Fable 5's reasoning while undercutting it on price.
The KeyNews take
Why it matters
Anthropic is competing on both capability and unit economics. A model that matches top-tier performance at half the token cost reshapes how enterprises evaluate LLM spend and vendor lock-in.
The key facts
5 to knowClaude Opus 5 scores 30.2% on ARC-AGI-3 benchmark
Performance nearly matches Fable 5 (benchmark comparison implied)
Token pricing at 50% of Fable 5 competitor
Outperforms GPT-5.6 Sol on novel problem-solving
Flagship model release targeting coding and knowledge work
Go to the source
The Decoderthe-decoder.com
Publisher excerpt: Anthropic's new flagship model Claude Opus 5 posts top scores in coding and knowledge work at half of Fable 5's token rates. On ARC-AGI-3, a benchmark for novel problem-solving, Opus 5 hits 30.2 percent, nearly four times higher than GPT-5.6 Sol.