Kimi K2.5vsQwen3.8-Max-0902

Kimi K2.5
Qwen3.8-Max-0902
Specifications
Parameters
1T2.4T
Context window
256k1M
API pricing
Input price
$0.60
Output price
$3.00
Cached input price
$0.10
Cheapest input
$0.45DeepInfra
Cheapest output
$2.25DeepInfra
Benchmarks
BullshitBench v2
52%
SWE-Bench Verified
76.8%
DeepSWE 1.1
69.3%
Next.js Evals
19%
NL2Repo-Bench
64.9%
QwenSWEBench V2
70%
LiveCodeBench
85%
Terminal-Bench 3.0
29%
JobBench
64%
CoWorkBench
76.1%
Toolathlon-Verified
73.3%
BrowseComp
60.6%
Humanity's Last Exam · with tools
30.1%
GPQA Diamond
87.6%
AutomationBench
50.8%
MMMU-Pro
82.7%
Overview
CompanyMoonshot AIQwen
Release dateJan 27 2026Sep 2 2026
AccessOpen WeightProprietary

Other comparisons

Kimi K2.5vsClaude Fable 5.1Qwen3.8-Max-0902vsClaude Fable 5.1Kimi K2.5vsGPT-5.6 SolQwen3.8-Max-0902vsGPT-5.6 SolKimi K2.5vsGemini 3.7 FlashQwen3.8-Max-0902vsGemini 3.7 FlashKimi K2.5vsMuse GlimmerQwen3.8-Max-0902vsMuse GlimmerKimi K2.5vsGrok 4.6Qwen3.8-Max-0902vsGrok 4.6Kimi K2.5vsDeepSeek-V4-Pro-0813Qwen3.8-Max-0902vsDeepSeek-V4-Pro-0813

Frequently asked questions

Kimi K2.5 and Qwen3.8-Max-0902 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Only Kimi K2.5 has a verified first-party API price: $0.60 per million input tokens and $3.00 per million output tokens. No pay-as-you-go API rate is tracked for Qwen3.8-Max-0902. Kimi K2.5 shipped 218 days before Qwen3.8-Max-0902, so benchmark comparisons should account for the intervening progress.

Kimi K2.5 has 1T parameters, while Qwen3.8-Max-0902 has 2.4T. Context windows are 256k (Kimi K2.5) vs 1M (Qwen3.8-Max-0902). Kimi K2.5 is open weight, while Qwen3.8-Max-0902 is proprietary.

Direct benchmark comparisons are unavailable — Kimi K2.5 and Qwen3.8-Max-0902 don't publish scores on any of the same benchmarks.

Kimi K2.5 was released by Moonshot AI on Jan 27 2026.

Qwen3.8-Max-0902 was released by Qwen on Sep 2 2026.

Only Kimi K2.5 has a verified first-party API price: $0.60 per million input tokens and $3.00 per million output tokens. No pay-as-you-go API rate is tracked for Qwen3.8-Max-0902. Rates are pay-as-you-go API prices verified on August 18, 2026.

Kimi K2.5 has a 256k context window; Qwen3.8-Max-0902 has 1M.

Kimi K2.5 is an open weight model released by Moonshot AI. Qwen3.8-Max-0902 is a proprietary model released by Qwen.