invokeable()
← comparisons / claude-opus-5-vs-gpt-5.6-luna

Claude Opus 5 vs GPT-5.6 Luna

Side-by-side from the Invokeable directory — provider-attributed pricing, canonical IDs, draft data flagged. Open as a shareable card →

01Head to head

Claude Opus 5GPT-5.6 Luna
CreatorAnthropicOpenAI
Released2026-07-242026-07-09
Context1M1.1M
Max output128K128K
Input $/M——
Output $/M——
Knowledge cutoff2026-052026-02
Modalitiestext + image → texttext + image → text
Capabilitiestool-use, extended-thinkingtool-use, extended-thinking, structured-output
ProvidersAnthropic APIOpenAI API

02Benchmarks · vendor-reported unless verified

Claude Opus 5GPT-5.6 Luna
BrowseComp (accuracy)—83.3
DeepSWE68.8%67.2%
FrontierCode (mean@5)53.4—
FrontierMath (accuracy)—78.6
GPQA Diamond (accuracy)—92.3
SWE-bench Multilingual89.5%—
SWE-bench Multimodal59.4%—
SWE-bench Pro79.2%62.7%
SWE-bench Verified (resolved)96—
Terminal-Bench (success rate)—84.7