Compare AI models

Specifications
Parameters
70B14.3B
Context window
128k32k
API pricing
Cheapest input
$0.10DeepInfra—
Cheapest output
$0.32DeepInfra—
$36.00DeepInfra · turbo · fp8—
Benchmarks
HumanEval
88.4%34.2%
MMLU
86%62.5%
Overview
CompanyMetaQwen
Release dateDec 6 2024Mar 28 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

LLaMA 3.3 leads Qwen1.5-MoE on 2 of the 2 benchmarks they both report (HumanEval, MMLU). Qwen1.5-MoE shipped 253 days before LLaMA 3.3, so benchmark comparisons should account for the intervening progress.

LLaMA 3.3 has 70B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 128k (LLaMA 3.3) vs 32k (Qwen1.5-MoE).

On HumanEval, LLaMA 3.3 leads at 88.4% vs Qwen1.5-MoE at 34.2%. On MMLU, LLaMA 3.3 leads at 86% vs Qwen1.5-MoE at 62.5%.