Compare AI models

Specifications
Parameters
—14B
Context window
128k2k
These models have no shared benchmark scores.
Overview
CompanyOpenAIQwen
Release dateFeb 12 2026Sep 25 2023
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

GPT-5.3-Codex-Spark and Qwen-14B don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Qwen-14B shipped 871 days before GPT-5.3-Codex-Spark, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (GPT-5.3-Codex-Spark) vs 2k (Qwen-14B). GPT-5.3-Codex-Spark is closed, while Qwen-14B is open weight.

Direct benchmark comparisons are unavailable — GPT-5.3-Codex-Spark and Qwen-14B don't publish scores on any of the same benchmarks.