Compare AI models

Specifications
Parameters
141B14B
Context window
64k2k
API pricing
Cheapest input
$2.00Mistral—
Cheapest output
$6.00Mistral—
$700Mistral · zdr—
Benchmarks
HumanEval
45.1%32.3%
MMLU
77.75%66.3%
GSM8K
78.6%61.3%
Overview
CompanyMistralQwen
Release dateApr 10 2024Sep 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mixtral 8×22B leads Qwen-14B on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Qwen-14B shipped 198 days before Mixtral 8×22B, so benchmark comparisons should account for the intervening progress.

Mixtral 8×22B has 141B parameters, while Qwen-14B has 14B. Context windows are 64k (Mixtral 8×22B) vs 2k (Qwen-14B).

On HumanEval, Mixtral 8×22B leads at 45.1% vs Qwen-14B at 32.3%. On MMLU, Mixtral 8×22B leads at 77.75% vs Qwen-14B at 66.3%. On GSM8K, Mixtral 8×22B leads at 78.6% vs Qwen-14B at 61.3%.