Compare AI models

Specifications
Parameters
1.05T—
Context window
1M1.05M
API pricing
Input price
$0.70$2.00
Output price
$2.10$12.00
Cached input price
$0.07$0.20
Cheapest input
$0.70Mistral$1.00OpenAI
Cheapest output
$2.10Mistral$6.00OpenAI
$245Mistral list price$1,000OpenAI list price
Benchmarks
BullshitBench v2
80%54%
DeepSWE 1.1
61.7%69.6%
Terminal-Bench 4.0
41%21.52%
BrowseComp
84.5%87.5%
AutomationBench
59.9%15.2%
threejseval
10081434
Overview
CompanyMistralOpenAI
Release dateOct 6 2026Jun 26 2026
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Mistral Large 4 and GPT-5.6 Terra are evenly matched across the 6 benchmarks they both report. Mistral Large 4 is cheaper on both input and output: $0.70 vs $2.00 per million input tokens, and $2.10 vs $12.00 per million output tokens. Figures are base-tier rates. GPT-5.6 Terra shipped 102 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-5.6 Terra). Mistral Large 4 is open weight, while GPT-5.6 Terra is closed.

On BullshitBench v2, Mistral Large 4 leads at 80% vs GPT-5.6 Terra at 54%. On DeepSWE 1.1, GPT-5.6 Terra leads at 69.6% vs Mistral Large 4 at 61.7%. On Terminal-Bench 4.0, Mistral Large 4 leads at 41% vs GPT-5.6 Terra at 21.52%. On BrowseComp, GPT-5.6 Terra leads at 87.5% vs Mistral Large 4 at 84.5%. On AutomationBench, Mistral Large 4 leads at 59.9% vs GPT-5.6 Terra at 15.2%. On threejseval, GPT-5.6 Terra leads at 1434 vs Mistral Large 4 at 1008.