Compare AI models

Specifications
Parameters
1T4B/30B/235B
Context window
256k256k
API pricing
Cheapest input
$0.49AtlasCloud—
Cheapest output
$2.50AtlasCloud—
$223AtlasCloud · int4—
Benchmarks
LiveCodeBench
85%74.1%
AIME 2025
96.1%92.3%
GPQA Diamond
87.6%81.1%
MMLU-Pro
87.1%84.4%
Overview
CompanyMoonshot AIQwen
Release dateJan 27 2026Jul 21 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Kimi K2.5 leads Qwen3-2507 on 4 of the 4 benchmarks they both report (LiveCodeBench, AIME 2025, GPQA Diamond, MMLU-Pro). Qwen3-2507 shipped 190 days before Kimi K2.5, so benchmark comparisons should account for the intervening progress.

Kimi K2.5 has 1T parameters, while Qwen3-2507 has 4B/30B/235B. Context windows are 256k (Kimi K2.5) vs 256k (Qwen3-2507).

On LiveCodeBench, Kimi K2.5 leads at 85% vs Qwen3-2507 at 74.1%. On AIME 2025, Kimi K2.5 leads at 96.1% vs Qwen3-2507 at 92.3%. On GPQA Diamond, Kimi K2.5 leads at 87.6% vs Qwen3-2507 at 81.1%. On MMLU-Pro, Kimi K2.5 leads at 87.1% vs Qwen3-2507 at 84.4%.