Claude Opus 5.5
Anthropic's flagship general-purpose model, leading agentic coding benchmarks at 40% less cost than Opus 5.
About
Claude Opus 5.5, released September 2026, is the first model in Anthropic's Claude 5.5 family and replaced Opus 5 as the default general-purpose model. Anthropic positions it at roughly Claude Fable 5.1 capability for most work at about 40% lower cost than Opus 5.
It uses an adjustable effort setting (default medium) and a 1M-token context window with no long-context surcharge.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Anthropic-reported:
- Terminal-Bench 4.0: 66.4%
- FrontierCode v1.1: 54.4%
- GDPval-AA v2.1: 1846 Elo
Independent: Artificial Analysis Intelligence Index 58 — the highest measured there.
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Established provider with transparent, detailed pricing and a well-documented migration story (breaking changes published). Benchmark leadership in agentic coding is corroborated by an independent evaluator (Artificial Analysis Index 58). Minor caveats: headline benchmarks are provider-reported at max effort, and the savings claim is measured at different effort settings.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 9.0/10
- Maintenance · recency, releases, issue load
- 9.0/10
- Openness · license + docs
- 5.1/10
- Manual review · sanity read + red-flag check
- 9.0/10
- · provider-tier estimate — thinner evidence (no public adoption metric)
- · released 5 days ago
- · manual review: max-effort benchmarks; agentic evals may flatter harness builders