Compare AI models

Specifications
Parameters
117B14.3B
Context window
128k32k
API pricing
Cheapest input
$0.03AkashML—
Cheapest output
$0.15Near AI—
$13.50Venice · int4—
Benchmarks
MMLU
90%62.5%
Overview
CompanyOpenAIQwen
Release dateAug 5 2025Mar 28 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

gpt-oss-120b leads Qwen1.5-MoE on 1 of the 1 benchmark they both report (MMLU). Qwen1.5-MoE shipped 495 days before gpt-oss-120b, so benchmark comparisons should account for the intervening progress.

gpt-oss-120b has 117B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 128k (gpt-oss-120b) vs 32k (Qwen1.5-MoE).

On MMLU, gpt-oss-120b leads at 90% vs Qwen1.5-MoE at 62.5%.