Mistral Medium 3.5vsKimi K2.5

Mistral Medium 3.5
Kimi K2.5
Specifications
Parameters
128B1T
Context window
256k256k
API pricing
Input price
$1.50$0.60
Output price
$7.50$3.00
Cached input price
$0.10
Cheapest input
$0.45SiliconFlow
Cheapest output
$2.25SiliconFlow
Benchmarks
SWE-Bench Verified
77.6%76.8%
Benchmarks
BullshitBench v2
52%
Next.js Evals
16%
LiveCodeBench
85%
BrowseComp
60.6%
Humanity's Last Exam · with tools
30.1%
GPQA Diamond
87.6%
Overview
CompanyMistralMoonshot AI
Release dateApr 29 2026Jan 27 2026
AccessOpen WeightOpen Weight

Other comparisons

Mistral Medium 3.5vsClaude Fable 5.1Kimi K2.5vsClaude Fable 5.1Mistral Medium 3.5vsGPT-6 AstraKimi K2.5vsGPT-6 AstraMistral Medium 3.5vsGemini 3.8 FlashKimi K2.5vsGemini 3.8 FlashMistral Medium 3.5vsMuse Spark 1.3Kimi K2.5vsMuse Spark 1.3Mistral Medium 3.5vsGrok 4.6Kimi K2.5vsGrok 4.6Mistral Medium 3.5vsDeepSeek-V4.1-FlashKimi K2.5vsDeepSeek-V4.1-Flash

Frequently asked questions

Mistral Medium 3.5 leads Kimi K2.5 on 1 of the 1 benchmark they both report (SWE-Bench Verified). Kimi K2.5 is cheaper on both input and output: $0.60 vs $1.50 per million input tokens, and $3.00 vs $7.50 per million output tokens. Kimi K2.5 shipped 92 days before Mistral Medium 3.5, so benchmark comparisons should account for the intervening progress.

Mistral Medium 3.5 has 128B parameters, while Kimi K2.5 has 1T. Context windows are 256k (Mistral Medium 3.5) vs 256k (Kimi K2.5).

On SWE-Bench Verified, Mistral Medium 3.5 leads at 77.6% vs Kimi K2.5 at 76.8%.