Compare AI models

Specifications
Parameters
—1.05T
Context window
1M1M
API pricing
Input price
$10.00$0.70
Output price
$50.00$2.10
Cached input price
$0.25$0.07
Cheapest input
$10.00Anthropic$0.70Mistral
Cheapest output
$50.00Anthropic$2.10Mistral
$4,500Anthropic list price$245Mistral list price
Benchmarks
BullshitBench v2
77%80%
DeepSWE 1.1
67.4%61.7%
FrontierCode 2.0
32.4%19.7%
Terminal-Bench 4.0
57.88%41%
AutomationBench
31.4%59.9%
AA-Briefcase v1.1
16781393
threejseval
18631008
Overview
CompanyAnthropicMistral
Release dateSep 1 2026Oct 6 2026
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude Fable 5.1 leads Mistral Large 4 on 5 of the 7 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. Claude Fable 5.1 shipped 35 days before Mistral Large 4, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Claude Fable 5.1) vs 1M (Mistral Large 4). Claude Fable 5.1 is closed, while Mistral Large 4 is open weight.

On BullshitBench v2, Mistral Large 4 leads at 80% vs Claude Fable 5.1 at 77%. On DeepSWE 1.1, Claude Fable 5.1 leads at 67.4% vs Mistral Large 4 at 61.7%. On FrontierCode 2.0, Claude Fable 5.1 leads at 32.4% vs Mistral Large 4 at 19.7%. On Terminal-Bench 4.0, Claude Fable 5.1 leads at 57.88% vs Mistral Large 4 at 41%. On AutomationBench, Mistral Large 4 leads at 59.9% vs Claude Fable 5.1 at 31.4%. On AA-Briefcase v1.1, Claude Fable 5.1 leads at 1678 vs Mistral Large 4 at 1393. On threejseval, Claude Fable 5.1 leads at 1863 vs Mistral Large 4 at 1008.