Compare AI models

Specifications
Parameters
109B>1T
Context window
10M256k
API pricing
Cheapest input
$0.08DeepInfra—
Cheapest output
$0.24DeepInfra—
$35.00DeepInfra · fp8—
Benchmarks
LiveCodeBench
32.8%85.9%
GPQA Diamond
57.2%87.4%
MMLU-Pro
74.3%85.7%
Overview
CompanyMetaQwen
Release dateApr 5 2025Jan 26 2026
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen3-Max-Thinking leads LLaMA 4 Scout on 3 of the 3 benchmarks they both report (LiveCodeBench, GPQA Diamond, MMLU-Pro). LLaMA 4 Scout shipped 296 days before Qwen3-Max-Thinking, so benchmark comparisons should account for the intervening progress.

LLaMA 4 Scout has 109B parameters, while Qwen3-Max-Thinking has >1T. Context windows are 10M (LLaMA 4 Scout) vs 256k (Qwen3-Max-Thinking). LLaMA 4 Scout is open weight, while Qwen3-Max-Thinking is closed.

On LiveCodeBench, Qwen3-Max-Thinking leads at 85.9% vs LLaMA 4 Scout at 32.8%. On GPQA Diamond, Qwen3-Max-Thinking leads at 87.4% vs LLaMA 4 Scout at 57.2%. On MMLU-Pro, Qwen3-Max-Thinking leads at 85.7% vs LLaMA 4 Scout at 74.3%.