Compare AI models

Specifications
Parameters
1T—
Context window
128k1M
API pricing
Cheapest input
$0.57Novita—
Cheapest output
$2.30Novita—
$229Novita · fp8—
Benchmarks
LiveCodeBench
53.7%37.8%
IFEval
89.8%76.3%
MMLU-Pro
81.1%64.5%
Overview
CompanyMoonshot AIQwen
Release dateJul 11 2025Nov 15 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Kimi K2 leads Qwen2.5-Turbo (1M) on 3 of the 3 benchmarks they both report (LiveCodeBench, IFEval, MMLU-Pro). Qwen2.5-Turbo (1M) shipped 238 days before Kimi K2, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Kimi K2) vs 1M (Qwen2.5-Turbo (1M)). Kimi K2 is open weight, while Qwen2.5-Turbo (1M) is closed.

On LiveCodeBench, Kimi K2 leads at 53.7% vs Qwen2.5-Turbo (1M) at 37.8%. On IFEval, Kimi K2 leads at 89.8% vs Qwen2.5-Turbo (1M) at 76.3%. On MMLU-Pro, Kimi K2 leads at 81.1% vs Qwen2.5-Turbo (1M) at 64.5%.