Compare AI models

Specifications
Parameters
—1.05T
Context window
1M1M
API pricing
Input price
$1.25$0.70
Output price
$4.25$2.10
Cached input price
$0.15$0.07
Cheapest input
$1.25Google$0.70Mistral
Cheapest output
$4.25Google$2.10Mistral
$463Meta list price$245Mistral list price
Benchmarks
DeepSWE 1.1
75.4%61.7%
SWEAtlas CodeBase QnA
59.4%59.4%
Terminal-Bench 4.0
14.55%41%
AutomationBench
49.6%59.9%
threejseval
14421008
Overview
CompanyMetaMistral
Release dateSep 2 2026Oct 6 2026
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Muse Spark 1.3 and Mistral Large 4 are evenly matched across the 5 benchmarks they both report. Mistral Large 4 is cheaper on both input and output: $0.70 vs $1.25 per million input tokens, and $2.10 vs $4.25 per million output tokens. Muse Spark 1.3 shipped 34 days before Mistral Large 4, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Muse Spark 1.3) vs 1M (Mistral Large 4). Muse Spark 1.3 is closed, while Mistral Large 4 is open weight.

On DeepSWE 1.1, Muse Spark 1.3 leads at 75.4% vs Mistral Large 4 at 61.7%. On SWEAtlas CodeBase QnA, both models score 59.4%. On Terminal-Bench 4.0, Mistral Large 4 leads at 41% vs Muse Spark 1.3 at 14.55%. On AutomationBench, Mistral Large 4 leads at 59.9% vs Muse Spark 1.3 at 49.6%. On threejseval, Muse Spark 1.3 leads at 1442 vs Mistral Large 4 at 1008.