← All models
Compare

Claude Opus 4.8 vs DeepSeek-V3.2

Add, remove, or swap models to compare them side by side.

Claude Opus 4.8DeepSeek-V3.2
Claude Opus 4.8Anthropic · 1 in the newsDeepSeek-V3.2DeepSeek · 1 in the news
Scores
Intelligence (ECI)157146
Coding9627
Math9136
Reasoning & Knowledge7626
Agentic & Tools9429
Specifications
DeveloperAnthropicDeepSeek
FamilyClaudeDeepSeek
ReleasedMay 28, 2026Dec 1, 2025
Parameters
AvailabilityAPI accessOpen weights (unrestricted)
Context window1M131K
Price — $/M input$5.00$0.23
Price — $/M output$25.00$0.34
Inputstext, image, filetext
Outputstexttext
Benchmarks
AIME 2024/202598%88%
APEX43%7%
ARC-AGI93%57%
ARC-AGI-272%4%
FrontierMath47%22%
FrontierMath Tier 431%2%
GPQA Diamond91%83%
SimpleBench65%53%
SimpleQA Verified40%28%
WebDev Arena15521286
WeirdML83%47%
Terminal-Bench40%

Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.

Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.