Compare AI models

Specifications
Parameters
675B0.8B/2B/4B/9B
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%81.7%
IFBench
37.8%64.5%
MMLU-Pro
80.9%82.5%
MMMU-Pro
54%70.1%
Overview
CompanyMistralQwen
Release dateDec 2 2025Mar 2 2026
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen3.5 Small leads Mistral Large 3 on 4 of the 4 benchmarks they both report (GPQA Diamond, IFBench, MMLU-Pro, MMMU-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 Qwen3.5 Small. Mistral Large 3 shipped 90 days before Qwen3.5 Small, so benchmark comparisons should account for the intervening progress.

Mistral Large 3 has 675B parameters, while Qwen3.5 Small has 0.8B/2B/4B/9B. Context windows are 256k (Mistral Large 3) vs 256k (Qwen3.5 Small).

On GPQA Diamond, Qwen3.5 Small leads at 81.7% vs Mistral Large 3 at 64.1%. On IFBench, Qwen3.5 Small leads at 64.5% vs Mistral Large 3 at 37.8%. On MMLU-Pro, Qwen3.5 Small leads at 82.5% vs Mistral Large 3 at 80.9%. On MMMU-Pro, Qwen3.5 Small leads at 70.1% vs Mistral Large 3 at 54%.