Claude Opus 5
Points and comments are a snapshot, not live.
Anthropic releases Claude Opus 5, matching near-Fable intelligence at half the cost.
Claude Opus 5 is now available, offering near-Fable 5 frontier intelligence at half the price. It achieves state-of-the-art results on coding benchmarks like Frontier-Bench v0.1 and CursorBench 3.2, and triples the next-best model on ARC-AGI 3. It matches Opus 4.8's cost while greatly improving performance across software engineering, scientific research, and business automation tasks. Opus 5 is the new default on Claude Max and the strongest model on Claude Pro. It also demonstrates improved alignment, scoring 2.3 on misaligned behavior, and remains behind Mythos 5 on risky cybersecurity capabilities.
What commenters are saying
Many commenters are intrigued by the cost-performance ratio, noting Opus 5 offers near-Fable capability at half the price. Some question why to use Fable if Opus is nearly as good and cheaper, while others argue Fable remains superior for complex planning and tasks where precision justifies higher cost. A few highlight Opus 5's improved alignment as a notable achievement. Some users remain skeptical, calling the benchmark comparisons slightly misleading since Opus trails Fable on some metrics but is highlighted as "best" for agentic coding. Overall, the thread centers on the trade-off between capability and cost.