Back
M

Mistral Large 3 2512

Mistral, France169/191 correctLast updated May 20, 2026Website

Mistral Large 3 2512 by Mistral achieved an overall score of 87.4% on Appwrite Arena, ranking #12 of 12 benchmarked models. The benchmark tests how well AI models understand Appwrite - the open-source backend platform for authentication, TablesDB, storage, functions, and more. This model answered 169 of 191 questions correctly across categories including Auth, TablesDB, Functions, Storage, and CLI. During the benchmark it consumed 1,320,851 tokens costing $0.6857, averaging 29.1 tokens per second over 14m 26s. Compare Mistral Large 3 2512 with other LLMs on the Appwrite Arena leaderboard.

Total tokens
1,320,851Input: 1,295,608 Output: 25,243
Total duration
14m 26s
Avg speed
29.1 tok/s
Total cost
$ 0.6857
Cost/1MOverallFoundationAuthTablesDBFunctionsStorageSitesMessagingRealtimeCLI
$0.50Input: $0.50/1M tokens
Output: $1.50/1M tokens		87.4%84.2%94%77.4%84.5%93.8%94.5%93%84.5%83.6%
Deterministic14/17 correct
AI-Judged95% average score