Compare AI models

Specifications
Parameters
—27B
Context window
400k256k
These models have no shared benchmark scores.
Overview
CompanyOpenAIQwen
Release dateSep 15 2025Apr 22 2026
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

GPT-5-Codex and Qwen3.6-27B don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. GPT-5-Codex shipped 219 days before Qwen3.6-27B, so benchmark comparisons should account for the intervening progress.

Context windows are 400k (GPT-5-Codex) vs 256k (Qwen3.6-27B). GPT-5-Codex is closed, while Qwen3.6-27B is open weight.

Direct benchmark comparisons are unavailable — GPT-5-Codex and Qwen3.6-27B don't publish scores on any of the same benchmarks.