Compare AI models

Specifications
Parameters
1T>1T
Context window
256k256k
API pricing
Input price
$0.95—
Output price
$4.00—
Cached input price
$0.16—
Cheapest input
$0.3895Baidu—
Cheapest output
$1.64Baidu—
$160Baidu · fp4—
Benchmarks
SWE-Bench Verified
80.2%75.3%
LiveCodeBench
89.6%85.9%
Humanity's Last Exam · with tools
54%49.8%
GPQA Diamond
90.5%87.4%
Overview
CompanyMoonshot AIQwen
Release dateApr 20 2026Jan 26 2026
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Kimi K2.6 leads Qwen3-Max-Thinking on 4 of the 4 benchmarks they both report (SWE-Bench Verified, LiveCodeBench, Humanity's Last Exam, GPQA Diamond). Only Kimi K2.6 has a verified first-party price for an available API: $0.95 per million input tokens and $4.00 per million output tokens. No currently available pay-as-you-go API rate is tracked for Qwen3-Max-Thinking. Qwen3-Max-Thinking shipped 84 days before Kimi K2.6, so benchmark comparisons should account for the intervening progress.

Kimi K2.6 has 1T parameters, while Qwen3-Max-Thinking has >1T. Context windows are 256k (Kimi K2.6) vs 256k (Qwen3-Max-Thinking). Kimi K2.6 is open weight, while Qwen3-Max-Thinking is closed.

On SWE-Bench Verified, Kimi K2.6 leads at 80.2% vs Qwen3-Max-Thinking at 75.3%. On LiveCodeBench, Kimi K2.6 leads at 89.6% vs Qwen3-Max-Thinking at 85.9%. On Humanity's Last Exam · with tools, Kimi K2.6 leads at 54% vs Qwen3-Max-Thinking at 49.8%. On GPQA Diamond, Kimi K2.6 leads at 90.5% vs Qwen3-Max-Thinking at 87.4%.