Compare AI models

Specifications
Parameters
675B>1T
Context window
256k256k
API pricing
Input price
$0.50—
Output price
$1.50—
Cached input price
$0.05—
Cheapest input
$0.50Mistral—
Cheapest output
$1.50Mistral—
$175Mistral list price—
Benchmarks
GPQA Diamond
64.1%87.4%
IFBench
37.8%70.9%
MMLU-Pro
80.9%85.7%
Overview
CompanyMistralQwen
Release dateDec 2 2025Jan 26 2026
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen3-Max-Thinking leads Mistral Large 3 on 3 of the 3 benchmarks they both report (GPQA Diamond, IFBench, MMLU-Pro). Only Mistral Large 3 has a verified first-party price for an available API: $0.50 per million input tokens and $1.50 per million output tokens. No currently available pay-as-you-go API rate is tracked for Qwen3-Max-Thinking. Mistral Large 3 shipped 55 days before Qwen3-Max-Thinking, so benchmark comparisons should account for the intervening progress.

Mistral Large 3 has 675B parameters, while Qwen3-Max-Thinking has >1T. Context windows are 256k (Mistral Large 3) vs 256k (Qwen3-Max-Thinking). Mistral Large 3 is open weight, while Qwen3-Max-Thinking is closed.

On GPQA Diamond, Qwen3-Max-Thinking leads at 87.4% vs Mistral Large 3 at 64.1%. On IFBench, Qwen3-Max-Thinking leads at 70.9% vs Mistral Large 3 at 37.8%. On MMLU-Pro, Qwen3-Max-Thinking leads at 85.7% vs Mistral Large 3 at 80.9%.