AI models

DeepSeek V4 Pro vs Kimi K3.1 Pro

Same criteria, side by side. Where the gap is small, it may be measurement noise.

DeepSeek V4 Pro

US$ 1.32 / US$ 3.96 peak hours

Input US$/1M / Output US$/1M · Official pricing

Kimi K3.1 Pro

US$ 2.00 / US$ 10.00

Input US$/1M / Output US$/1M · Official pricing

GPQA Diamond

90.9% · 94.2%

DeepSeek V4 Pro · Kimi K3.1 Pro

01

Direct answer

Direct answer: GPQA Diamond: DeepSeek V4 Pro 90.9%, Kimi K3.1 Pro 94.2% (Epoch AI). FrontierMath (tiers 1 to 3): DeepSeek V4 Pro 45.3%, Kimi K3.1 Pro 78.2% (Epoch AI). SimpleQA Verified: DeepSeek V4 Pro 47%, Kimi K3.1 Pro 62.2% (Epoch AI). τ-bench Verified (airline): DeepSeek V4 Pro 78.6%, Kimi K3.1 Pro 78.7% (OpenRouter Benchmarks).
see more

02

Pick Kimi K3.1 Pro if scores come first: it leads on GPQA Diamond, FrontierMath and SimpleQA Verified and is in a statistical tie with DeepSeek V4 Pro on τ-bench Verified. Pick DeepSeek V4 Pro if cost matters more: at official prices it charges US$ 1.32 versus US$ 2.00 per 1M input tokens and US$ 3.96 versus US$ 10.00 per 1M output.

Kimi K3.1 Pro

  • GPQA Diamond +3.3 p.p.
  • FrontierMath +32.9 p.p.
  • SimpleQA Verified +15.2 p.p.

statistical tie

  • τ-bench Verified statistical tie

03

MetricDeepSeek V4 ProKimi K3.1 ProSource
Input US$/1MUS$ 1.32 / US$ 3.96 Official pricing peak hoursUS$ 2.00 / US$ 10.00 Official pricing
Output US$/1MUS$ 3.96US$ 10.00
Context1,048,5761,048,576
BENCHMARKS
GPQA Diamond90.9%between 34th and 60th of 250 94.2%between 9th and 26th of 250 Epoch AI
OTIS Mock AIME saturated96.7% statistical tie with the top score100% top scoreEpoch AI
FrontierMath (tiers 1 to 3)45.3%between 56th and 58th of 85 78.2%between 22nd and 24th of 85 Epoch AI
SimpleQA Verified47%between 36th and 41st of 87 62.2%between 17th and 18th of 87 Epoch AI
τ-bench Verified (airline)78.6%between 6th and 23rd of 130 78.7%between 6th and 22nd of 130 OpenRouter Benchmarks
OTIS Mock AIME saturated96.7% statistical tie with the top score100% top scoreEpoch AI

04

Compare with

05

Sources

  • Epoch AI, CC BY 4.0, latest version we checked.
  • OpenRouter Benchmarks, CC BY 4.0, latest version we checked.
  • OpenRouter, dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.

How to cite

If you use this data, cite the page and the date you accessed it.

Adriano Meirinho. "DeepSeek V4 Pro vs Kimi K3.1 Pro". meirinho.com.br. https://meirinho.com.br/en/ia/modelos/comparar/deepseek-v4-pro-vs-kimi-k3-1-pro/. Accessed November 21, 2026.

Third-party data follows the license of each source. The full table is available as CSV and JSON.

Markdown