← All models
Compare

DeepSeek-V4-Pro-0813 vs Kimi K3

DeepSeek-V4-Pro-0813 (DeepSeek) and Kimi K3 (Moonshot) compared on benchmarks, pricing, context window, and use-case rankings.

AttributeDeepSeek-V4-Pro-0813DeepSeek · 1 in the newsKimi K3Moonshot · 33 in the news
Scores
Intelligence (ECI)157
Coding8099
Math9082
Reasoning & Knowledge7878
Agentic & Tools84
Specifications
DeveloperDeepSeekMoonshot
FamilyDeepSeekKimi
ReleasedAug 13, 2026Jul 16, 2026
Parameters1.6T2.8T
AvailabilityOpen weights (unrestricted)Open weights (non-commercial)
Context window1M1M
Price — $/M input$0.66$2.80
Price — $/M output$1.98$14.00
Inputstexttext, image, video
Outputstexttext
Benchmarks
AIME 2024/202599%97%
GPQA Diamond92%93%
Humanity's Last Exam41%47%
SciCode49%59%
SimpleQA Verified55%43%
WebDev Arena15841682
WeirdML66%83%
APEX39%
ARC-AGI95%
ARC-AGI-260%

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.

Frequently asked questions

Which is cheaper, DeepSeek-V4-Pro-0813 or Kimi K3?

DeepSeek-V4-Pro-0813 is cheaper on input tokens at $0.66 per million, versus $2.80 (representative OpenRouter pricing).

Which has a larger context window, DeepSeek-V4-Pro-0813 or Kimi K3?

DeepSeek-V4-Pro-0813 supports up to 1M tokens, compared with 1M for the other.

Which is better for coding, DeepSeek-V4-Pro-0813 or Kimi K3?

Across coding benchmarks like SWE-bench Verified and Terminal-Bench, Kimi K3 ranks higher — 99th vs 80th percentile among the models tracked on Model Beat.

Want a different match-up? Open the compare tool to add or swap models.

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