Compare AI models

Specifications
Parameters
—0.5B/1.5B/3B/7B/14B/32B
Context window
8k128k
API pricing
Cheapest input
$30.00Azure$0.66Cloudflare
Cheapest output
$60.00Azure$1.00Cloudflare
$9,000Azure$182Cloudflare
Benchmarks
HumanEval
67%92.7%
MMLU
86.4%77.6%
GSM8K
92%93%
Overview
CompanyOpenAIQwen
Release dateMar 14 2023Sep 19 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Coder leads GPT-4 on 2 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). GPT-4 shipped 555 days before Qwen2.5-Coder, so benchmark comparisons should account for the intervening progress.

Context windows are 8k (GPT-4) vs 128k (Qwen2.5-Coder). GPT-4 is closed, while Qwen2.5-Coder is open weight.

On HumanEval, Qwen2.5-Coder leads at 92.7% vs GPT-4 at 67%. On MMLU, GPT-4 leads at 86.4% vs Qwen2.5-Coder at 77.6%. On GSM8K, Qwen2.5-Coder leads at 93% vs GPT-4 at 92%.