Venice Atlas
← Back to model catalog

VENICE ATLAS · MODEL DOSSIER · CLAUDE OPUS 5.5

Model overview

Claude Opus 5.5

Claude Opus 5.5 is a step-change improvement over Opus 5, with the biggest gains in agentic coding, long-running agentic tasks, and knowledge work. It is also a much better collaborator on long-running work, with a 1M token context window, 128K max output tokens, adaptive thinking, and strong multimodal capabilities.

claude-opus-5-5 · venice.ai
Text · Venice catalog$4.00 per 1M tokensPrice on Venice AI
ModalityTextOn Venice AI
Primary price$4.00 per 1M tokensper 1M tokens
Context window1MMax tokens
Benchmarks31Benchmark results

Benchmark results

See the tests behind each score. They use the benchmark’s original scoring scale.

Agents#145.42 ratingCoding#149.72 ratingVision#141.57 ratingGeneral intelligence#160.33 ratingHealthcare#1731.58 ratingMath#541.85 ratingMultimodal#141.16 ratingProductivity#215.49 ratingReasoning#159.33 ratingScience#418.73 ratingTool Calling#333.84 rating
Agents6 results
AA-Briefcase v1.1aa-briefcase-1.1 · agents
60.7%
ReportedOct 2026
AutomationBenchautomationbench · agents
40.0%
ReportedOct 2026
OfficeQAofficeqa · agents
78.9%
ReportedOct 2026
OfficeQA Proofficeqa-pro · agents
67.7%
ReportedOct 2026
Program Benchprogram-bench · agents
91.2%
ReportedOct 2026
Toolathlon-Verifiedtoolathlon-verified · agents
77.8%
ReportedOct 2026
Biology2 results
BioMysteryBenchbiomysterybench · biology
89.3%
ReportedOct 2026
LatchBio SpatialBench Verifiedlatchbio-spatialbench-verified · biology
72.0%
ReportedOct 2026
Coding5 results
BenchCADbenchcad · code
73.0%
ReportedOct 2026
BenchCAD (with Python tool)benchcad-with-python-tool · code
96.2%
ReportedOct 2026
CursorBench 4.0cursorbench-4.0 · code
57.8%
ReportedOct 2026
DeepSWE 1.1deepswe-1.1 · code
74.2%
ReportedOct 2026
FrontierSWE V2frontierswe-v2 · code
62.3%
ReportedOct 2026
General intelligence2 results
GDPval-AA 2.1gdpval-aa-2.1 · general
61.5%
ReportedOct 2026
Global-MMLUglobal-mmlu · general
94.3%
ReportedOct 2026
Healthcare2 results
HealthBenchhealthbench · healthcare
60.6%
ReportedOct 2026
HealthBench Professionalhealthbench-professional · healthcare
65.6%
ReportedOct 2026
Language1 result
MILUmilu · language
93.1%
ReportedOct 2026
Legal1 result
Legal Agent Benchmarklegal-agent-benchmark · legal
8.30%
ReportedOct 2026
Math1 result
ArXivMatharxivmath · math
91.2%
ReportedOct 2026
Multimodal1 result
OSWorld 2.0osworld-2.0 · multimodal
81.8%
ReportedOct 2026
Reasoning8 results
FrontierCode 1.1frontiercode-1.1 · reasoning
54.4%
ReportedOct 2026
Humanity's Last Exam (no tools, text-only)humanity's-last-exam-(no-tools,-text-only) · reasoning
64.4%
ReportedOct 2026
Humanity's Last Exam (with tools, text-only)humanity's-last-exam-(with-tools,-text-only) · reasoning
67.7%
ReportedOct 2026
SWE-bench Multilingualswe-bench-multilingual · reasoning
93.9%
ReportedOct 2026
SWE-Bench Multimodalswe-bench-multimodal · reasoning
61.4%
ReportedOct 2026
SWE-Bench Proswe-bench-pro · reasoning
89.9%
ReportedOct 2026
Terminal-Bench 4.0terminal-bench-4.0 · reasoning
66.4%
ReportedOct 2026
Terminal-Bench-Science 0.1terminal-bench-science-0.1 · reasoning
58.7%
ReportedOct 2026
Science1 result
LatchBio SingleCellBenchlatchbio-singlecellbench · science
61.2%
ReportedOct 2026

Learn what each benchmark measures ↗

Related models

More from anthropic

claude-sonnet-5-5Claude Sonnet 5.5$2.00 input per 1M tokens ↗claude-haiku-5-5Claude Haiku 5.5$0.10 input per 1M tokens ↗claude-fable-5-1Claude Fable 5.1$10.00 input per 1M tokens ↗