Compare AI models

Specifications
Parameters
1T>1T
Context window
256k256k
API pricing
Cheapest input
$0.60Novita—
Cheapest output
$2.50Novita—
$245Novita · fp8—
Benchmarks
SWE-Bench Verified
69.2%75.3%
Humanity's Last Exam · text only, no tools
7.9%30.2%
GPQA Diamond
74.2%87.4%
MMLU-Pro
81.9%85.7%
Overview
CompanyMoonshot AIQwen
Release dateSep 5 2025Jan 26 2026
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen3-Max-Thinking leads Kimi K2 (0905) on 4 of the 4 benchmarks they both report (SWE-Bench Verified, Humanity's Last Exam, GPQA Diamond, MMLU-Pro). Kimi K2 (0905) shipped 143 days before Qwen3-Max-Thinking, so benchmark comparisons should account for the intervening progress.

Kimi K2 (0905) has 1T parameters, while Qwen3-Max-Thinking has >1T. Context windows are 256k (Kimi K2 (0905)) vs 256k (Qwen3-Max-Thinking). Kimi K2 (0905) is open weight, while Qwen3-Max-Thinking is closed.

On SWE-Bench Verified, Qwen3-Max-Thinking leads at 75.3% vs Kimi K2 (0905) at 69.2%. On Humanity's Last Exam · text only, no tools, Qwen3-Max-Thinking leads at 30.2% vs Kimi K2 (0905) at 7.9%. On GPQA Diamond, Qwen3-Max-Thinking leads at 87.4% vs Kimi K2 (0905) at 74.2%. On MMLU-Pro, Qwen3-Max-Thinking leads at 85.7% vs Kimi K2 (0905) at 81.9%.