When was Mistral Medium 3.5 released?

Mistral Medium 3.5 was released by Mistral on Apr 29 2026.

When was Kimi K2 Thinking released?

Kimi K2 Thinking was released by Moonshot AI on Nov 6 2025.

Which is better at coding, Mistral Medium 3.5 or Kimi K2 Thinking?

On SWE-Bench Verified (real-world software-engineering tasks), Mistral Medium 3.5 leads — Mistral Medium 3.5 scores 77.6% and Kimi K2 Thinking scores 71.3%.

How do the context windows of Mistral Medium 3.5 and Kimi K2 Thinking compare?

Mistral Medium 3.5 has a 256k context window; Kimi K2 Thinking has 256k.

Home Analytics Contact

Latest Release

Mistral Medium 3.5

Apr 29 2026

Home/Compare/Mistral Medium 3.5 vs Kimi K2 Thinking

Mistral Medium 3.5 vs Kimi K2 Thinking

1 vs 0 benchmarks won

	Mistral Mistral Medium 3.5	Moonshot AI Kimi K2 Thinking
Overview
Company	Mistral	Moonshot AI
Release date	Apr 29 2026	Nov 6 2025
Model type	—	—
Open source	Yes	Yes
Specifications
Parameters	128B	1T
Context window	256k	256k
Benchmarks
Science reasoning GPQA Diamond	—	—
Software engineering SWE-Bench Verified	77.6%	71.3%
Multimodal understanding MMMU	—	—
Timeline
Release gap	Kimi K2 Thinking shipped 174 days before Mistral Medium 3.5

Which is better: Mistral Medium 3.5 or Kimi K2 Thinking?

Mistral Medium 3.5 leads Kimi K2 Thinking on 1 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). Kimi K2 Thinking shipped 174 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 Thinking has 1T. Context windows are 256k (Mistral Medium 3.5) vs 256k (Kimi K2 Thinking).

On SWE-Bench Verified, Mistral Medium 3.5 scores 77.6%, 6.3 points above Kimi K2 Thinking at 71.3%.

Frequently asked questions

When was Mistral Medium 3.5 released?: Mistral Medium 3.5 was released by Mistral on Apr 29 2026.
When was Kimi K2 Thinking released?: Kimi K2 Thinking was released by Moonshot AI on Nov 6 2025.
Which is better at coding, Mistral Medium 3.5 or Kimi K2 Thinking?: On SWE-Bench Verified (real-world software-engineering tasks), Mistral Medium 3.5 leads — Mistral Medium 3.5 scores 77.6% and Kimi K2 Thinking scores 71.3%.
How do the context windows of Mistral Medium 3.5 and Kimi K2 Thinking compare?: Mistral Medium 3.5 has a 256k context window; Kimi K2 Thinking has 256k.