Compare AI models

Specifications
Parameters
8B/70B/405B14B
Context window
128k2k
API pricing
Cheapest input
$0.02DeepInfra—
Cheapest output
$0.04DeepInfra—
$6.00DeepInfra · fp8—
Benchmarks
HumanEval
89%32.3%
MATH
73.8%24.8%
MMLU
87.3%66.3%
GSM8K
96.8%61.3%
Overview
CompanyMetaQwen
Release dateJul 23 2024Sep 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

LLaMA 3.1 leads Qwen-14B on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). Qwen-14B shipped 302 days before LLaMA 3.1, so benchmark comparisons should account for the intervening progress.

LLaMA 3.1 has 8B/70B/405B parameters, while Qwen-14B has 14B. Context windows are 128k (LLaMA 3.1) vs 2k (Qwen-14B).

On HumanEval, LLaMA 3.1 leads at 89% vs Qwen-14B at 32.3%. On MATH, LLaMA 3.1 leads at 73.8% vs Qwen-14B at 24.8%. On MMLU, LLaMA 3.1 leads at 87.3% vs Qwen-14B at 66.3%. On GSM8K, LLaMA 3.1 leads at 96.8% vs Qwen-14B at 61.3%.