Compare AI models

Specifications
Parameters
1.05T—
Context window
1M1.05M
API pricing
Input price
$0.70$10.00
Output price
$2.10$50.00
Cached input price
$0.07$1.00
Cheapest input
$0.70Mistral$5.00OpenAI
Cheapest output
$2.10Mistral$25.00OpenAI
$245Mistral list price$4,500OpenAI list price
Benchmarks
BullshitBench v2
80%69%
DeepSWE 1.1
61.7%74.1%
FrontierCode 2.0
19.7%42.8%
AA Coding Agent Index
4867
Terminal-Bench 4.0
41%57.9%
BrowseComp
84.5%91.5%
AutomationBench
59.9%41.4%
AA-Briefcase v1.1
13931569
threejseval
10081889
Overview
CompanyMistralOpenAI
Release dateOct 6 2026Sep 3 2026
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GPT-6 Astra leads Mistral Large 4 on 7 of the 9 benchmarks they both report. Mistral Large 4 is cheaper on both input and output: $0.70 vs $10.00 per million input tokens, and $2.10 vs $50.00 per million output tokens. Figures are base-tier rates. GPT-6 Astra shipped 33 days before Mistral Large 4, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Mistral Large 4) vs 1.05M (GPT-6 Astra). Mistral Large 4 is open weight, while GPT-6 Astra is closed.

On BullshitBench v2, Mistral Large 4 leads at 80% vs GPT-6 Astra at 69%. On DeepSWE 1.1, GPT-6 Astra leads at 74.1% vs Mistral Large 4 at 61.7%. On FrontierCode 2.0, GPT-6 Astra leads at 42.8% vs Mistral Large 4 at 19.7%. On AA Coding Agent Index, GPT-6 Astra leads at 67 vs Mistral Large 4 at 48. On Terminal-Bench 4.0, GPT-6 Astra leads at 57.9% vs Mistral Large 4 at 41%. On BrowseComp, GPT-6 Astra leads at 91.5% vs Mistral Large 4 at 84.5%. On AutomationBench, Mistral Large 4 leads at 59.9% vs GPT-6 Astra at 41.4%. On AA-Briefcase v1.1, GPT-6 Astra leads at 1569 vs Mistral Large 4 at 1393. On threejseval, GPT-6 Astra leads at 1889 vs Mistral Large 4 at 1008.