From the experiment

Do coding agents recommend Mistral web search?

Mistral web search was chosen in 2% of 499 judged AI search sessions, ranking tenth. Measured with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code.

Published September 23, 2026 Read as Markdown

Mistral web search was chosen in 2% of 499 judged AI search sessions, ranking tenth. It was also raised as a candidate in 9 further sessions without being chosen.

This page reports what happened when Claude Code, Codex, Cursor, Grok Build CLI and Muse Code had to solve a problem in AI search inside a realistic codebase. Not what a chat assistant says about Mistral web search. What an agent actually installed.

One thing to read first: every one of those wins came from a single codebase. That is a result about one repository rather than about AI search in general, and the splits below cannot separate the two. Treat them as a description of that repository.

The numbers

CategoryAI search
Sessions in the category499
Sessions where Mistral web search was chosen12
Install share2%
Rank in category10 of 41
Codebases it won in1
Raised as a candidate, not chosen9
Chosen when considered57%
Sitemistral.ai

By agent

With 12 wins spread across 5 agents, the rates below are small numbers and a difference between them is not yet a finding. They are here because the direction is worth knowing, not because the gap is established.

AgentSessionsChose Mistral web searchShare
Claude Code14054%
Codex14453%
Cursor13611%
Grok Build CLI3513%
Muse Code4400%

By who was asking

Mistral web search performs similarly across the four kinds of buyer, from 0% to 4%. That is unusual: the category leader changed with the persona in 23 of the 29 categories we measured.

Who is askingSessionsChose Mistral web searchShare
Vibe coder8200%
Junior developer17374%
Senior engineer12400%
Enterprise team12054%

What Mistral web search was up against

The full ranking in AI search, from the same sessions:

#ProductRuns wonShare
1Anthropic web search8417%
2Brave Search API6914%
3Exa6713%
4Tavily428%
5OpenAI web search275%
6Built in-house (no product adopted)245%
7Firecrawl235%
8NewsAPI.ai235%

What this means

Agents almost never raise Mistral web search without choosing it: 9 sessions raised against 12 chosen. When it gets considered, it usually wins. The constraint is how rarely it gets considered at all, which is a presence problem: templates, framework integrations and pages that answer the configuration and production questions agents actually search for.

Where these numbers come from

The 499 sessions in AI search are part of a published set of 11,878, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.

Every AI search run can be replayed on the board.

If you work on Mistral web search: the judge recorded a reason for every session where it was raised and passed over. Those reasons are in the transcripts.

<!-- generated by scripts/write-data-pages.mjs -->

Common questions

Do coding agents recommend Mistral web search?

Yes. Mistral web search was chosen in 12 of the 499 judged sessions in AI search, a 2% install share, ranking tenth in its category.

Does Claude Code recommend Mistral web search?

In 5 of the 140 sessions in AI search run with Claude Code, which is 4%.

Do different coding agents treat Mistral web search differently?

Not much. Claude Code, Codex, Cursor, Grok Build CLI and Muse Code chose it at similar rates, between 0% and 4% of their runs.

How was this measured?

Real coding agents at pinned versions were run in sandboxes inside 91 realistic codebases and asked to solve real tasks. A simulated project owner approved or questioned each recommendation before any code was written, and a judge from a model family that builds none of the agents read every session blind.

How often is Mistral web search considered but not chosen?

It was raised as a candidate in 9 sessions without being chosen, and chosen in 12. That is a 57% conversion from considered to chosen.

Where this comes from

Armature ran 11,878 judged sessions with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code inside 91 realistic codebases, and published every run. The numbers on this page come from that work.

Read next

All library pages