LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|
Toolofficial publisher

web_search (Mistral)

Tool : A precise action the AI can call · requires mistral-api

Web search for Mistral agents, premium version with AFP and AP.

Sovereign and sourced, but $30 per 1,000 calls.

Overall score
6.5 /10
Security: one caveat, see the card
Utility
7.0
Reliability
6.0
Security
One caveat
Ease
1 click · Comfortable

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

AI agents need access to recent information that their base model does not contain. Mistral's web_search tool allows an agent to run real-time web searches directly through the Mistral API, without integrating a third-party service.

In practice, the tool is declared in the tools list when creating an agent or calling the Conversations API. When the agent determines that a search is needed, it executes it server-side on Mistral's infrastructure and incorporates the results into its response. The premium version (web_search_premium) relies on professional news sources such as AFP and AP.

This tool is aimed at developers building agents or automated workflows via the Mistral API who need fresh, verifiable, and sourced data. It is particularly suited to journalistic use cases, monitoring, or news summarization.

It is not suitable for users working exclusively with the standard Chat Completions API, as web_search is not compatible with that endpoint. It is also not adapted for those looking for a free or low-cost tool for intensive use.

What it offers

  • ·Real-time web search natively integrated into Mistral agents, with no external configuration required.
  • ·Server-side execution on Mistral's infrastructure: no local code to manage for running searches.
  • ·Premium version with access to professional AFP and AP sources for more reliable results (paid).
  • ·Compatible with the Conversations API and the Agents API, usable directly without creating an agent first.
  • ·Ability to combine multiple tools simultaneously within the same agent (web_search, code_interpreter, image_generation, etc.).
  • ·Also available to end users via the Conversations interface, without going through agent creation.

In practice

Price
Paid · from 28 €
Commercial use
Commercial use: yes
Available on
api
Host
mistral-api
FR interface
No

Security and data

Rating
Security: one caveat, see the card
Capabilities
readswritesexecutesnetwork
Your data
Data with the vendor
Maintenance
Active

Links and repository checked on 25 October 2026 · added on 1 May 2025

How to install web_search (Mistral)

1. Create an account on the Mistral platform and obtain an API key from your developer dashboard. 2. Install the Mistral SDK in your environment (Python or TypeScript depending on your stack). 3. When creating an agent via client.beta.agents.create(), add web_search or web_search_premium to the tools parameter. 4. Use only the Conversations API (/v1/conversations) or the Agents API: web_search is not supported by the Chat Completions API. 5. Keep your API key confidential and never commit it to your code repository.
tools=[{"type":"web_search"}]

Official documentation ↗

In the same category

Whole category Feed (memory, context) →