Compare AI models

Specifications
Parameters
27B/35B/122B27B
Context window
256k262k
API pricing
Cheapest input
—$0.035Darkbloom
Cheapest output
—$1.05StreamLake
—$70.30Chutes · fp8
Benchmarks
LiveCodeBench
78.9%90.3%
Humanity's Last Exam · no tools
25.3%30.8%
GPQA Diamond
86.6%89.2%
IFBench
76.1%79.5%
OSWorld-Verified
58%84.3%
Overview
CompanyQwenQwen
Release dateFeb 24 2026Aug 14 2026
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen3.8-27B leads Qwen3.5 Medium on 5 of the 5 benchmarks they both report. Qwen3.5 Medium shipped 171 days before Qwen3.8-27B, so benchmark comparisons should account for the intervening progress.

Qwen3.5 Medium has 27B/35B/122B parameters, while Qwen3.8-27B has 27B. Context windows are 256k (Qwen3.5 Medium) vs 262k (Qwen3.8-27B).

On LiveCodeBench, Qwen3.8-27B leads at 90.3% vs Qwen3.5 Medium at 78.9%. On Humanity's Last Exam · no tools, Qwen3.8-27B leads at 30.8% vs Qwen3.5 Medium at 25.3%. On GPQA Diamond, Qwen3.8-27B leads at 89.2% vs Qwen3.5 Medium at 86.6%. On IFBench, Qwen3.8-27B leads at 79.5% vs Qwen3.5 Medium at 76.1%. On OSWorld-Verified, Qwen3.8-27B leads at 84.3% vs Qwen3.5 Medium at 58%.