The Open-Source AI Stack
RSS
All models

Models · Compare

Claude Opus 4 vs DeepSeek-R1 (May 2025 refresh)

Rows highlighted in warm gray are where the models differ. Numbers carry their as-of date and primary source.

Specs

Field A: Claude Opus 4 B: DeepSeek-R1 (May 2025 refresh)
Released 2025-05-222025-05-28
Developer AnthropicDeepSeek
Openness ProprietaryOpen
License ProprietaryMIT
OSI-approved noyes
Data released nono
Training code nono
Architecture unknownmoe
Total params
Active params
Experts
Context window
Attention unknownmla
Position enc. unknownrope-yarn
Pretraining tokens
Post-training rlhf, constitutionalsft, grpo, rejection-sampling
Training hardware H800
$/M input $15.00
$/M output $75.00
Output tok/sec 38.1

Benchmarks

Missing scores render as not reported; never inferred. Bold highlights the leader per benchmark.

General reasoning

MMLU-Pro 86.0 2026-05-21 85.0 2025-05-28
GPQA-Diamond 74.9 2025-05-22 81.0 2025-05-28

Code

SWE-Bench Verified 72.5 2025-05-22
LiveCodeBench 54.2 2026-05-21 73.3 2025-05-28

Math

MATH 94.1 2026-05-21
AIME 2024 56.3 2026-05-21 91.4 2025-05-28
AIME 2025 33.9 2025-05-22

Context · A

Flagship of the May 22 2025 Claude 4 launch. Returned to the Opus name after the Claude 3.5 and 3.7 lines skipped Opus entirely. Held the same $15 / $75 input/output price ceiling as Claude 3 Opus, and like Sonnet 4 carried hybrid reasoning with extended thinking and parallel tool execution.

Context · B

An RL-only refresh of R1 that gained substantial ground on reasoning benchmarks (notably AIME 2024) without any new pretraining. Tightened the open-vs-closed reasoning gap.

Claude Opus 4 detail → · DeepSeek-R1 (May 2025 refresh) detail →