ModelsMulti-sourceSelected

Mistral Large 4 tops AutomationBench business workflow benchmark

Why we picked it

Agent benchmarks that measure office-app workflows, rather than chat or coding, are becoming a separate leaderboard from traditional evals, and Mistral is staking its position on it. Developers building automation agents on Gmail, Sheets or Salesforce should check whether the benchmark's task design matches their use cases before switching models. Comparisons against Kimi K3 and DeepSeek V4 Pro also frame it as a European lab competing on agentic execution.

Mistral AI announced on X that Mistral Large 4 leads its AutomationBench benchmark, which tests 657 business workflows across simulated workplace apps including Gmail, Google Sheets, Slack and Salesforce.

Body

The model finished ahead of Kimi K3, MiMo-V2.6-Pro and DeepSeek V4 Pro on the benchmark.

The results were published via Mistral AI's official X account with no additional detail in the post.

Original · Mistral AI

Gmail. Google Sheets. Slack. Salesforce. i.e. the four horsemen of Monday. On AutomationBench, Mistral Large 4 took on 657 business workflows across simulated workplace apps, all four included, finishing ahead of Kimi K3, MiMo-V2.6-Pro and DeepSeek V4 Pro.📊 8 ⚡

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Models" desk, focusing on Europe's AI governance and technology ecosystem. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison and AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Mistral Large 4, AutomationBench, DeepSeek, Kimi.

Source attribution: curated from Mistral AI (@MistralAI). Cross-source corroboration: 11 independent sources have covered this development (including Artificial Analysis, lmarena.ai, Guillaume Lample (Mistral)), confirming industry-wide signal strength.

  • Artificial Analysis10-06 13:45Source
  • lmarena.ai10-07 19:45Source
  • Guillaume Lample (Mistral)10-06 13:24Source
  • TestingCatalog10-06 13:28Source
  • Julien Chaumond10-06 13:30Source
  • OpenRouter10-06 14:02Source
  • Gorden Sun10-06 14:02Source
  • DeepLearning.AI10-08 16:00Source
  • Arthur Mensch10-06 13:29Source
  • IT之家10-06 13:56Source