invokeable()
← directory / models / claude-opus-5

Claude Opus 5

model claude-opus-5

Draft entry — numbers not yet verified against primary sources.

1M
context
128K
max output
—
input
—
output
2026-05
knowledge cutoff
1
providers

01Providers

ProviderProvider model IDIn $/MOut $/MContextStatus
Anthropic API claude-opus-5 — — 1M live

02Invoke

provider
language

  

  



03Benchmarks · vendor-reported unless verified
SWE-bench Verified (resolved) resolved96




SWE-bench Pro resolve rate79.2%




SWE-bench Multilingual resolve rate89.5%




SWE-bench Multimodal resolve rate59.4%




DeepSWE resolve rate68.8%




FrontierCode (mean@5) mean@5 · Main53.4




04Related
Claude Opus 4.82026-05Claude Opus 4.72026-04Claude Opus 4.62026-02Claude Opus 4.52025-11Claude Opus 4.12025-08Claude Opus 42025-05


05Compare
vs GPT-5.6 Lunahead-to-headvs Qwen3.8 27Bhead-to-headvs Gemini 3.7 Flashhead-to-headvs Mistral Medium 3.5head-to-headvs Nemotron 3.5 Lightning 30B A3Bhead-to-headvs GLM-5.3head-to-headvs North Mini Codehead-to-headvs Claude Sonnet 5head-to-headvs Claude Opus 4.8head-to-head


06Sources · where these numbers come from
  1. https://github.com/sst/models.dev
  2. https://docs.claude.com/en/docs/about-claude/models
  3. https://www.anthropic.com/news
  4. https://docs.claude.com