Compare AI models

Specifications
Parameters
1.05T—
Context window
1M400k
API pricing
Input price
$0.70$1.75
Output price
$2.10$14.00
Cached input price
$0.07$0.175
Cheapest input
$0.70Mistral$1.75Azure
Cheapest output
$2.10Mistral$14.00Azure
$245Mistral list price$1,050OpenAI list price
Benchmarks
BullshitBench v2
80%24%
threejseval
10081153
Overview
CompanyMistralOpenAI
Release dateOct 6 2026Feb 5 2026
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Mistral Large 4 and GPT-5.3-Codex are evenly matched across the 2 benchmarks they both report (BullshitBench v2, threejseval). Mistral Large 4 is cheaper on both input and output: $0.70 vs $1.75 per million input tokens, and $2.10 vs $14.00 per million output tokens. GPT-5.3-Codex shipped 243 days before Mistral Large 4, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Mistral Large 4) vs 400k (GPT-5.3-Codex). Mistral Large 4 is open weight, while GPT-5.3-Codex is closed.

On BullshitBench v2, Mistral Large 4 leads at 80% vs GPT-5.3-Codex at 24%. On threejseval, GPT-5.3-Codex leads at 1153 vs Mistral Large 4 at 1008.