Compare AI models

Specifications
Parameters
—1.05T
Context window
1M1M
API pricing
Input price
$0.30$0.70
Output price
$2.50$2.10
Cached input price
$0.03$0.07
Cheapest input
$0.15Google AI Studio$0.70Mistral
Cheapest output
$1.25Google AI Studio$2.10Mistral
$185Google list price$245Mistral list price
Benchmarks
BullshitBench v2
19%80%
Overview
CompanyGoogleMistral
Release dateApr 17 2025Oct 6 2026
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral Large 4 leads Gemini 2.5 Flash on 1 of the 1 benchmark they both report (BullshitBench v2). Gemini 2.5 Flash is cheaper on input: $0.30 vs $0.70 per million tokens. Mistral Large 4 is cheaper on output: $2.10 vs $2.50 per million tokens. Gemini 2.5 Flash shipped 537 days before Mistral Large 4, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Gemini 2.5 Flash) vs 1M (Mistral Large 4). Gemini 2.5 Flash is closed, while Mistral Large 4 is open weight.

On BullshitBench v2, Mistral Large 4 leads at 80% vs Gemini 2.5 Flash at 19%.