invokeable()
← comparisons / claude-opus-5-vs-glm-5.1

Claude Opus 5 vs GLM-5.1

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 5GLM-5.1
CreatorAnthropicZhipuai
Released2026-07-242026-04-07
Context1M200K
Max output128K131K
Input $/M—$0.98/M
Output $/M—$3.08/M
Knowledge cutoff2026-05—
Modalitiestext + image → texttext → text
Capabilitiestool-use, extended-thinkingtool-use, extended-thinking, structured-output
ProvidersAnthropic APIOpenRouter

02Benchmarks · vendor-reported unless verified

Claude Opus 5GLM-5.1
Artificial Analysis Coding Agent Index (average pass@1)—52.7
DeepSWE68.8%—
FrontierCode (mean@5)53.4—
SWE-Atlas Codebase QnA (pass@1)—73.2
SWE-bench Multilingual89.5%—
SWE-bench Multimodal59.4%—
SWE-bench Pro79.2%—
SWE-bench Pro (pass@1)—19.8
SWE-bench Verified (resolved)96—
Terminal-Bench (pass@1)—65.1