LLM ReferenceLLM Reference
All comparisons

DeepSeek V3 vs Claude Opus 4.5

Side-by-side comparison of specifications, capabilities, and pricing.

Released2024-12-262025-11-01
Context window64k200K
Parameters671B
Architecturemixture of expertsdecoder only
LicenseOpen SourceProprietary
Knowledge cutoff2024-042025-12

Capabilities

Vision
Multimodal
Reasoning
Function calling
Tool use
Structured Outputs
Code execution

Availability

Providers

Benchmarks

HellaSwag95.7—HumanEval85.5—Massive Multitask Language Understanding88.5—LiveCodeBench49.6—Aider Polyglot48.4—BigCodeBench50.0—Chatbot Arena1302.0—MMLU PRO75.988.9Massive Multi-discipline Multimodal Understanding—80.7BFCL—77.5SWE-bench Pro—41.8