DeepSeek V4 Pro 0813 or Kimi K2.5: which to choose?

The match-up at a glance

DeepSeek V4 Pro 0813 and Kimi K2.5 are two open-weight models that you can install on your own servers, one from DeepSeek, the other from Moonshot AI. DeepSeek V4 Pro 0813 was released on August 13, 2026, Kimi K2.5 on January 27, 2026.

All figures below come from public sources checked on October 22, 2026: models.dev and OpenRouter for prices and capabilities, Epoch AI for scores.

What the benchmarks say

On our capability index, DeepSeek V4 Pro 0813 scores 72.2 out of 100 and Kimi K2.5 78.1. Kimi K2.5 leads by 5.9 points.

On the 2 benchmarks reported for both, DeepSeek V4 Pro 0813 wins 2 times and Kimi K2.5 0:

• GPQA Diamond (PhD-level science questions): DeepSeek V4 Pro 0813 91.7%, Kimi K2.5 87.6%
• OTIS Mock AIME (competition mathematics problems): DeepSeek V4 Pro 0813 98.6%, Kimi K2.5 92.2%

Pricing

DeepSeek V4 Pro 0813 costs $0.66 per million input tokens and $1.98 per million output tokens; Kimi K2.5 costs $0.58 and $3.

For a typical agent task (read 20,000 tokens, write 2,000), expect $0.017 with DeepSeek V4 Pro 0813 versus $0.018 with Kimi K2.5. The cost is almost the same.

Run the numbers with your own volumes in the agent cost calculator.

Under the hood

DeepSeek V4 Pro 0813 reads more in one go: 1,048,576 tokens of context (roughly 1,573 pages), versus 262,144 for Kimi K2.5.

Reasoning mode: DeepSeek V4 Pro 0813 yes, Kimi K2.5 yes. Tool calling: DeepSeek V4 Pro 0813 yes, Kimi K2.5 yes. Image input: DeepSeek V4 Pro 0813 no, Kimi K2.5 yes.

Both publish their weights: you can host them yourself.

In short

Our reading of the data:

• Choose DeepSeek V4 Pro 0813 for a larger context window and a more recent release.
• Choose Kimi K2.5 for a higher capability index and image input.

This verdict relies on public figures; it does not replace a trial on your own tasks.