Overview
Summary
The useful answer before the supporting detail.
Mistral Vibe is an AI chat assistant and agent for work and code, formerly known as Le Chat S7. It is used by professionals and developers to automate tasks, write and deploy code, and create documents and presentations S7. Its practical value lies in integrating with 100+ tools and offering flexible deployment, including on-premises for enterprises S7.
Product identity and positioning
Name, category, positioning, audience, and first-party claims.
- Identity: Mistral Vibe is an AI chat assistant and agent for work and code, formerly Le Chat S7.
- Core purpose: It solves the problem of repetitive work and complex coding tasks by automating them S7.
- Intended users: Professionals and developers who need AI assistance for work and coding S7.
- Product scope: It is a single application offering chat, coding, research, and document creation, not a platform S7.
- Positioning: Mistral positions Vibe as an AI agent for long-horizon tasks, fluent in your knowledge and tools S7.
- Claimed distinction: It claims to handle long-horizon tasks with deep integration and customization S7.
Use cases and audiences
Who the product serves and the jobs it is designed to handle.
| Audience | Practical use case |
|---|---|
| Developers | Write, test, and deploy code with async agents that open PRs S7. |
| Business professionals | Generate board decks, strategic briefs, and reports from research S7. |
| Teams | Automate recurring workflows and schedule tasks S7. |
| Enterprises | Deploy on-premises with full data residency for compliance S7. |
Traffic and search
Search summary
The useful answer before the supporting detail.
Frontier open-weight models you own and control. Customize with Forge, deploy with Studio, run on-prem, your VPC, or our cloud. mistral.ai: an estimated 6,984,878 total monthly visits and about 58,615 monthly organic visits across 11,497 ranking keywords. Scout crawled 15 pages and measured a technical health score of 94/100.
- Total traffic (modeled): 6,984,878 monthly visits across all channels.
- Organic search traffic (estimated): 58,615 monthly visits from search.
- Strongest sampled ranking: devstral at position 1.
- Leading page in the collected evidence: https://mistral.ai/.
- Closest measured organic competitor: ollama.com, appearing 3 times in the sampled first-page SERPs.
- Technical crawl: 15 pages, health score 94/100.
Website intelligence — Collected evidence
Company identity and positioning
| Page type | Evidence | Source |
|---|---|---|
| other | Frontier open-weight models you own and control. Customize with Forge, deploy with Studio, run on-prem, your VPC, or our cloud. | https://mistral.ai/ |
| other | // AFTER initial render. Astro's ClientRouter swaps / on // navigation, detaching those injected nodes: the banner disappears and // openAxeptioCookies() stops working (SDK DOM refs point to detached // nodes). A full reload re-runs GTM → re-injects Axeptio → works again. // // Fix: tag every node injected post-render, park the exact same nodes // under (outside head/body) during the swap, then move them back // after. Nodes stay in the same document throughout — no cross-document // adoption — so the Axeptio SDK's JS references remain valid. Scripts // are skipped so the SDK is never re-executed (no duplicate banner). var PERSIST = 'data-consent-persist' var PARK = 'data-consent-park' var observer = null function tag(node) { if (!node \ | node.nodeType !== 1) return if (node.tagName === 'SCRIPT') return if (node.hasAttribute(PERSIST) \ |
| other | All customers | https://mistral.ai/ |
| other | ](/customers/) | https://mistral.ai/ |
| other | Run on-prem, in your VPC, or have us run it for you. Keep inference in-region and reserve dedicated GPU capacity for predictable throughput without managing hardware. | https://mistral.ai/ |
| other | Learn about our customers | https://mistral.ai/ |
| other | “By combining a century of geoscience data, the expertise of our teams and Mistral's capabilities, we aim to develop a new generation of tools capable of supporting our experts in analyzing the most complex data and in their decision-making.” | https://mistral.ai/ |
| other | “Thanks to the high-performance models and made-to-measure support of Mistral experts, we are building the foundations necessary to power our current and future products and services, enabling us to serve our customers better.” | https://mistral.ai/ |
| features | // AFTER initial render. Astro's ClientRouter swaps / on // navigation, detaching those injected nodes: the banner disappears and // openAxeptioCookies() stops working (SDK DOM refs point to detached // nodes). A full reload re-runs GTM → re-injects Axeptio → works again. // // Fix: tag every node injected post-render, park the exact same nodes // under (outside head/body) during the swap, then move them back // after. Nodes stay in the same document throughout — no cross-document // adoption — so the Axeptio SDK's JS references remain valid. Scripts // are skipped so the SDK is never re-executed (no duplicate banner). var PERSIST = 'data-consent-persist' var PARK = 'data-consent-park' var observer = null function tag(node) { if (!node \ | node.nodeType !== 1) return if (node.tagName === 'SCRIPT') return if (node.hasAttribute(PERSIST) \ |
| features | All customers | https://mistral.ai/solutions/coding/ |
| features | ](/customers/) | https://mistral.ai/solutions/coding/ |
| features | Transform how your teams build software. | https://mistral.ai/solutions/coding/ |
Domain performance
Modeled total traffic, six-month trend, engagement, countries, and domain identity.
Total website traffic
| Month | Modeled visits |
|---|---|
| 2026-04 | 9,642,927 |
| 2026-05 | 9,741,026 |
| 2026-06 | 9,251,575 |
| 2026-07 | 7,843,106 |
| 2026-08 | 6,954,249 |
| 2026-09 | 6,984,878 |
Top visitor countries
| Country | Name | Traffic share |
|---|---|---|
| FR | France | 38.8% |
| DE | Germany | 14.2% |
| US | United States | 6.5% |
| NL | Netherlands | 4.5% |
| IN | India | 4.0% |
Domain identity
| Signal | Value |
|---|---|
| Registered | May 15, 2019 |
| Last changed | Aug 11, 2026 |
| Expires | May 15, 2029 |
| WHOIS server | whois.nic.ai |
| DNSSEC | Not signed |
| Nameservers | ivan.ns.cloudflare.com, ada.ns.cloudflare.com |
Website capture

Organic keywords
Top keywords and an independent Google result page.
Top organic keywords
| Keyword | Volume | CPC | Estimated value |
|---|---|---|---|
| mistral | 441,720 | $1.84 | $384,390 |
| mistral ai | 206,860 | $1.07 | $203,550 |
| le chat | 61,590 | $1.14 | $65,420 |
| lechat | 28,780 | $1.15 | $35,290 |
| mistral le chat | 27,740 | $0.81 | $28,320 |
Independent Google result page
Query: devstral
| Position | Title | URL | Description |
|---|---|---|---|
| 1 | Devstral | Mistral AI | https://mistral.ai/news/devstral/ | The most powerful AI platform for enterprises. Devstral, our agentic LLM for software engineering tasks. to solve real GitHub issues; |
| 2 | devstral | https://ollama.com/library/devstral | Devstral is an agentic LLM for software engineering tasks built. Devstral excels at using tools to explore codebases, editing multiple files and power software ... |
| 3 | mistralai/Devstral-Small-2505 | https://huggingface.co/mistralai/Devstral-Small-2505 | Devstral is an agentic LLM for software engineering tasks built under a collaboration between Mistral AI and All Hands AI |
| 4 | Introducing: Devstral 2 and Mistral Vibe CLI. | https://mistral.ai/news/devstral-2-vibe-cli/ | The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with ... |
| 5 | Meet Mistral Devstral, SOTA open model designed ... | https://www.reddit.com/r/LocalLLaMA/comments/1kryxdg/meet_mistral_devstral_sota_open_model_designed/ | Meet Mistral Devstral, SOTA open model designed specifically for coding agents … devstral small is faster and better than glm 4.7 flash for ... |
| 6 | Devstral: Fine-tuning Language Models for Coding Agent ... | https://arxiv.org/pdf/2509.25193 | by A Rastogi · 2025 · Cited by 27 — We introduce Devstral-Small, a lightweight open source model for code agents with the best performance among models below 100B size. |
| 7 | Developing with Devstral is Both Delightful and Disappointing | https://www.youtube.com/watch?v=LB0z43mxTcg | Devstral, a new coding LLM from Mistral AI. It is designed to tackle real GitHub issues and complex software engineering tasks. |
| 8 | Devstral | https://lmstudio.ai/models/devstral | Devstral is a coding model from Mistral AI. It excels at using tools to explore codebases, editing multiple files and power software engineering agents. |
| 9 | Devstral 2 : Best Small Coding LLMs by Mistral | https://medium.com/data-science-in-your-pocket/devstral-2-best-small-coding-llms-by-mistral-32821d5c5db8 | Devstral 2 is one of the clearest signals that open-source coding models are catching up in the areas that matter most: reliability, agent ... |
| 10 | Upgrading agentic coding capabilities with the new Devstral ... | https://mistral.ai/news/devstral-2507/ | The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with ... |
Important pages
Pages earning the most estimated search traffic in the sample.
| URL | Ranking keywords | Estimated traffic | Best position |
|---|---|---|---|
| https://mistral.ai/ | 7 | 361.8 | 1 |
| https://mistral.ai/news/devstral/ | 2 | 276.6 | 1 |
| https://mistral.ai/news/codestral/ | 1 | 179.4 | 1 |
| https://mistral.ai/news/devstral-2-vibe-cli/ | 1 | 179.4 | 1 |
| https://mistral.ai/news/le-chat-mistral/ | 2 | 170.2 | 1 |
| https://mistral.ai/products/vibe/ | 2 | 130.7 | 1 |
| https://mistral.ai/fr/ | 1 | 63.8 | 1 |
| https://docs.mistral.ai/models | 1 | 9.1 | 1 |
| https://mistral.ai/news/codestral-mamba/ | 1 | 9.1 | 1 |
| https://mistral.ai/news/codestral-25-08/ | 1 | 6.1 | 1 |
Website intelligence — Collected evidence
Prioritized website intelligence pages
| Type | Page | Words | Change | URL |
|---|---|---|---|---|
| other | Mistral \ | static | 809 | new |
| pricing | Pricing \ | static | 1,422 | new |
| features | Coding \ | static | 810 | new |
| features | Document AI and OCR \ | static | 2,729 | new |
| features | Speech intelligence for enterprises \ | static | 1,355 | new |
| customers | Customer stories \ | static | 833 | new |
| customers | Airbus \ | static | 433 | new |
| customers | Ardian \ | static | 1,196 | new |
| customers | ASML and Mistral. \ | static | 434 | new |
| customers | Austrian Academy of Sciences \ | static | 957 | new |
| customers | AXA revolutionizes the future of insurance with AI. \ | static | 415 | new |
| customers | Belfius \ | static | 494 | new |
| customers | BigDataCorp \ | static | 487 | new |
| customers | BMW \ | static | 431 | new |
Features
| Page type | Evidence | Source |
|---|---|---|
| features | // AFTER initial render. Astro's ClientRouter swaps / on // navigation, detaching those injected nodes: the banner disappears and // openAxeptioCookies() stops working (SDK DOM refs point to detached // nodes). A full reload re-runs GTM → re-injects Axeptio → works again. // // Fix: tag every node injected post-render, park the exact same nodes // under (outside head/body) during the swap, then move them back // after. Nodes stay in the same document throughout — no cross-document // adoption — so the Axeptio SDK's JS references remain valid. Scripts // are skipped so the SDK is never re-executed (no duplicate banner). var PERSIST = 'data-consent-persist' var PARK = 'data-consent-park' var observer = null function tag(node) { if (!node \ | node.nodeType !== 1) return if (node.tagName === 'SCRIPT') return if (node.hasAttribute(PERSIST) \ |
| features | Get in touch | https://mistral.ai/solutions/coding/ |
| features | ](/contact/) | https://mistral.ai/solutions/coding/ |
| features | ](/research/) | https://mistral.ai/solutions/coding/ |
| features | Sovereign AI | https://mistral.ai/solutions/coding/ |
| features | ](/contact)[ | https://mistral.ai/solutions/coding/ |
| features | ](https://console.mistral.ai) | https://mistral.ai/solutions/coding/ |
| features | Why Sovereign AI | https://mistral.ai/solutions/coding/ |
| features | ](/sovereign-ai/)[ | https://mistral.ai/solutions/coding/ |
| features | ](/products/)[ | https://mistral.ai/solutions/coding/ |
| features | ](/our-method/)[ | https://mistral.ai/solutions/coding/ |
| features | Sovereign AI in practice | https://mistral.ai/solutions/coding/ |
| features | ](/use-cases/) | https://mistral.ai/solutions/coding/ |
| features | ](/cloud/inference/)[ | https://mistral.ai/solutions/coding/ |
| features | ](/cloud/compute/) | https://mistral.ai/solutions/coding/ |
| features | Energy & utilities | https://mistral.ai/solutions/coding/ |
| features | ](/industry/energy/)[ | https://mistral.ai/solutions/coding/ |
| features | Manufacturing | https://mistral.ai/solutions/coding/ |
| features | ](/industry/manufacturing/)[ | https://mistral.ai/solutions/coding/ |
| features | Financial services | https://mistral.ai/solutions/coding/ |
Pricing and plans
| Page type | Evidence | Source |
|---|---|---|
| pricing | Education plan | https://mistral.ai/pricing/ |
| pricing | $25.5 /mo in API credits. | https://mistral.ai/pricing/ |
| pricing | $15 /mo in API credits. | https://mistral.ai/pricing/ |
| pricing | ](https://mistral.ai/pricing/enterprise-deployments/) | https://mistral.ai/pricing/ |
| pricing | ](https://mistral.ai/pricing/enterprise-deployments/) \ | static |
| pricing | Up to 6x free | https://mistral.ai/pricing/ |
| pricing | Up to 5x free | https://mistral.ai/pricing/ |
| pricing | Up to 40x free | https://mistral.ai/pricing/ |
| pricing | Keep going past your plan limit by paying only for what you use beyond it at API rate. | https://mistral.ai/pricing/ |
| pricing | Does Mistral have a free plan? | https://mistral.ai/pricing/ |
| pricing | Yes, Mistral’s Free plan provides limited access to Vibe, including our latest models, chat, search, creation, and Vibe for code for agentic coding in your terminal. On the Free plan, you also get access to Mistral Studio. | https://mistral.ai/pricing/ |
| pricing | How is API pricing calculated? | https://mistral.ai/pricing/ |
| pricing | Most models are priced per million tokens, with both input (your prompts) and output (responses) counted separately. For example, Mistral Large costs $0.5 /M tokens in and $1.5 /M tokens out. Batch processing, for high-volume work, reduces the price by 50%, and cached input tokens reduce input cost by up to 90% for repeated prompts. A few APIs are priced differently: OCR is per 1,000 pages, speech models are per minute, and tool APIs are priced per call. | https://mistral.ai/pricing/ |
| pricing | What’s the difference between Mistral Free and Pro plans? | https://mistral.ai/pricing/ |
| pricing | The Free plan is ideal for getting started with the Vibe or Studio. Mistral Pro offers higher limits across chat plus full access to coding with Mistral Vibe. You get more messages and web searches, more extended thinking and deep research reports, expanded Mistral Vibe capacity, and chat and email support. | https://mistral.ai/pricing/ |
| pricing | Pro has higher ceilings compared to Free (e.g. up to 40x more image generations). Team and Enterprise plans have even higher allowances on certain features. For example, Flash answers: 200 / day on Team vs. 150 / day on Pro. | https://mistral.ai/pricing/ |
| pricing | Verified students can get Mistral Pro for $5.99 / month (normally $14.99). | https://mistral.ai/pricing/ |
| pricing | Yes, the Mistral Enterprise plan includes custom SLAs, dedicated support, and private deployments. Contact us for more details. | https://mistral.ai/pricing/ |
Integrations and supported platforms
No supporting public-page evidence was collected for this category.
Product and company updates
No supporting public-page evidence was collected for this category.
Rank tracking
Persistent keyword baselines and position movement.
This analysis created the first position baseline at 2026-10-18T04:29:46.878Z. Movement appears after the next fresh analysis using the same location and language.
| Keyword | Current | Previous | Movement | Volume | Ranking URL |
|---|---|---|---|---|---|
| devstral | 1 | Baseline | Baseline | 880 | https://mistral.ai/news/devstral/ |
| codestral | 1 | Baseline | Baseline | 590 | https://mistral.ai/news/codestral/ |
| devstral 2 | 1 | Baseline | Baseline | 590 | https://mistral.ai/news/devstral-2-vibe-cli/ |
| french ai | 1 | Baseline | Baseline | 590 | https://mistral.ai/ |
| chat le | 1 | Baseline | Baseline | 390 | https://mistral.ai/news/le-chat-mistral/ |
| chat mistral | 1 | Baseline | Baseline | 260 | https://mistral.ai/products/vibe/ |
| french ai company | 1 | Baseline | Baseline | 260 | https://mistral.ai/ |
| ia en francais | 1 | Baseline | Baseline | 210 | https://mistral.ai/fr/ |
| chat mistral ai | 1 | Baseline | Baseline | 170 | https://mistral.ai/products/vibe/ |
| french le chat | 1 | Baseline | Baseline | 170 | https://mistral.ai/news/le-chat-mistral/ |
| ai mistral | 1 | Baseline | Baseline | 90 | https://mistral.ai/ |
| france ai | 1 | Baseline | Baseline | 90 | https://mistral.ai/ |
| french llm | 1 | Baseline | Baseline | 70 | https://mistral.ai/ |
| french ai companies | 1 | Baseline | Baseline | 50 | https://mistral.ai/ |
| french ai model | 1 | Baseline | Baseline | 40 | https://mistral.ai/ |
| best mistral model | 1 | Baseline | Baseline | 30 | https://docs.mistral.ai/models |
| codestral mamba | 1 | Baseline | Baseline | 30 | https://mistral.ai/news/codestral-mamba/ |
| devstral model | 1 | Baseline | Baseline | 30 | https://mistral.ai/news/devstral/ |
| codestral 2508 | 1 | Baseline | Baseline | 20 | https://mistral.ai/news/codestral-25-08/ |
| devstral medium | 1 | Baseline | Baseline | 20 | https://docs.mistral.ai/models/model-cards/devstral-medium-1-0-25-07 |
Features
Features and capabilities
Core functions and evidence-backed product capabilities.
| Capability | What it does |
|---|---|
| Chat and search | Ask questions and get answers with web search and internal knowledge sources S7. |
| Coding agent | Write, test, and deploy code across your codebase, with async agents that open PRs S7. |
| Document creation | Generate PDFs, board decks, briefs, and reports from research and data S7. |
| Workflow automation | Schedule recurring tasks and package workflows into reusable agents S7. |
| Data analysis | Upload spreadsheets or connect databases to spot patterns and flag anomalies S7. |
| Integrations | Connect to 100+ tools via connectors and MCP compatibility S7. |
| Enterprise deployment | Deploy on-premises, in private cloud, or on Mistral Cloud with data residency S7. |
Integrations and ecosystem
Connected products, exports, APIs, and workflow compatibility.
| Integration or interface | Type | What it enables |
|---|---|---|
| Email, calendar, Slack, GitHub, Jira | Connectors | Connect to 100+ tools for search and action S7. |
| MCP servers | MCP compatibility | Connect to any remote MCP server S73. |
| VS Code and JetBrains | Plugin | Code completion and natural language editing S73. |
| External knowledge sources | Connectors | Connect sources for real-time answers S81. |
| API | API | Programmatic access to models and features S103. |
Pricing
Pricing and plans
Published plans, free access, trials, credits, and pricing limitations.
| Plan or offer | Verified price and billing | Included | Limits |
|---|---|---|---|
| Free | $0 S73 | Limited messages, web searches, coding sessions S73 | Limited libraries; up to 5 scheduled tasks S73 |
| Pro | $14.99 per month, excluding taxes S73 | $25.5/mo in API credits S73 | Up to 5x free web searches; up to 15 GB libraries S73 |
| Team | $24.99 per user per month, excluding taxes S73 | Up to 5x free web searches S73 | Up to 30 GB per user; $50 minimum for 2 users S73 |
| Enterprise | Contact sales S73 | Private deployments, audit logs, SAML SSO S73 | Not published S73 |
| Student | $5.99 per month (normally $14.99) S73 | Pro features S73 | Valid for 12 months for verified students S73 |
Competition
Competitive landscape
Domains competing for the same organic search demand.
Sampled competitors
| Competitor | First-page appearances | Sampled queries | Average position | Queries observed |
|---|---|---|---|---|
| ollama.com | 3 | 2 | 6 | devstral, codestral |
| datacamp.com | 2 | 2 | 4 | devstral, codestral |
| medium.com | 2 | 2 | 4.5 | devstral, codestral |
| youtube.com | 2 | 1 | 4.5 | devstral |
| reddit.com | 2 | 2 | 7.5 | devstral, codestral |
| digitalocean.com | 1 | 1 | 2 | devstral |
| huggingface.co | 1 | 1 | 3 | codestral |
| venturebeat.com | 1 | 1 | 5 | devstral |
| lmstudio.ai | 1 | 1 | 6 | codestral |
Website intelligence — Collected evidence
Customers, testimonials, and first-party review claims
| Page type | Evidence | Source |
|---|---|---|
| customers | // AFTER initial render. Astro's ClientRouter swaps / on // navigation, detaching those injected nodes: the banner disappears and // openAxeptioCookies() stops working (SDK DOM refs point to detached // nodes). A full reload re-runs GTM → re-injects Axeptio → works again. // // Fix: tag every node injected post-render, park the exact same nodes // under (outside head/body) during the swap, then move them back // after. Nodes stay in the same document throughout — no cross-document // adoption — so the Axeptio SDK's JS references remain valid. Scripts // are skipped so the SDK is never re-executed (no duplicate banner). var PERSIST = 'data-consent-persist' var PARK = 'data-consent-park' var observer = null function tag(node) { if (!node \ | node.nodeType !== 1) return if (node.tagName === 'SCRIPT') return if (node.hasAttribute(PERSIST) \ |
| customers | Get in touch | https://mistral.ai/customers/ |
| customers | ](/contact/) | https://mistral.ai/customers/ |
| customers | ](/research/) | https://mistral.ai/customers/ |
| customers | Sovereign AI | https://mistral.ai/customers/ |
| customers | ](/contact)[ | https://mistral.ai/customers/ |
| customers | ](https://console.mistral.ai) | https://mistral.ai/customers/ |
| customers | Why Sovereign AI | https://mistral.ai/customers/ |
| customers | ](/sovereign-ai/)[ | https://mistral.ai/customers/ |
| customers | ](/products/)[ | https://mistral.ai/customers/ |
| customers | ](/our-method/)[ | https://mistral.ai/customers/ |
| customers | Sovereign AI in practice | https://mistral.ai/customers/ |
| customers | ](/use-cases/) | https://mistral.ai/customers/ |
| customers | ](/cloud/inference/)[ | https://mistral.ai/customers/ |
| customers | ](/cloud/compute/) | https://mistral.ai/customers/ |
| customers | Energy & utilities | https://mistral.ai/customers/ |
| customers | ](/industry/energy/)[ | https://mistral.ai/customers/ |
| customers | Manufacturing | https://mistral.ai/customers/ |
| customers | ](/industry/manufacturing/)[ | https://mistral.ai/customers/ |
| customers | Financial services | https://mistral.ai/customers/ |
Independent review sources discovered
No links to supported independent review platforms were discovered on the collected pages.
Backlink profile
Authority and referring-domain signals.
Backlink comparison profile
- Domain rating: 88/100
- Backlinks: 44,739,368
- Referring domains: 42,252
- Dofollow backlinks: 86.0%
- Dofollow referring domains: 82.0%
Collection decision
The Scout search index remains intentionally excluded from automatic Scout analysis. This Scout traffic model profile costs 5 credits (0.5 cents) and is included in the first comparison run so its usefulness can be judged before the recurring collection is finalized.
AI visibility
Measured ChatGPT and Google AI mentions with cited sources.
AI-assistant referral traffic
- Estimated current generative-AI referral share: 4.1%
| Month | Traffic share | Source |
|---|---|---|
| — | — | All AI assistants |
| — | — | All AI assistants |
| — | — | All AI assistants |
Website intelligence — Collected evidence
Site-stated AI capabilities
These are first-party product claims, not independent measurements of AI-search visibility.
Marketing opportunities
Actions calculated directly from the collected evidence.
| Opportunity | Recommended action | Evidence |
|---|---|---|
| Competitor gap | Review the sampled queries where ollama.com appears. | 3 first-page appearances across devstral, codestral |
Website intelligence — Collected evidence
Website evidence gaps
- integrations: no usable public page was collected; verify whether the site publishes this evidence.
- reviews: no usable public page was collected; verify whether the site publishes this evidence.
- changes: no usable public page was collected; verify whether the site publishes this evidence.
- identity: no usable public page was collected; verify whether the site publishes this evidence.
- technical: no usable public page was collected; verify whether the site publishes this evidence.
Trust and proof
Customers, proof, and reviews
Published proof, customer signals, and review evidence.
Independent reviews
No citation-grounded independent review links were found.
Customer proof
| Proof type | Named customer or source | Evidence |
|---|---|---|
| Case study | Cisco | customer case study — # Transforming the CX experience: How Cisco plans to cut renewal proposal time by 20%. (Cisco / Mistral AI) |
| Case study | Belfius | customer case study — Belfius, a leading Northern European Digital Bank, aimed to enhance their customer experience by migrating from a pure scripted bot platform to an advanced AI solution. (Belfius / Mistral AI) |
| Case study | Pretto | customer case study — Pretto, the leading online French mortgage broker, leverages Mistral models to improve the experience of their borrower clients by analyzing in depth their communications with their broker. (Pretto / Mistral AI) |
Independent mentions
- Customer stories / Mistral — # Bringing customers to the frontier. We help customers across the world transform their operations, products, and services with AI. ... semiconductor lithography with Mistral. ## AXA empowers employees and develops tailored use cases with Mistral.
- Zalando / Mistral AI — , a leading European e-commerce platform, is dedicated to delivering exceptional shopping experiences to its millions of customers. By partnering with AWS ... engagement and operational efficiency. Zalando utilized Mistral models hosted on AWS
- Stellantis and Mistral / Mistral AI — vehicles to streamlining production, AI is a key piece of ... is working with Mistral to develop and evaluate AI-driven innovations such as a Bill of Materials Data Intelligence tool. This tool is designed to streamline the analysis of complex component
- Ardian / Mistral — of unstructured documents daily—from market and due ... memoranda. To accelerate deal execution and enhance decision-making capabilities, Ardian partnered with Artefact and Mistral to develop GAIA, an internal Generative AI platform that transforms how
- Capgemini / Mistral AI — % to 100% developer adoption, the internal coding ... - 100% developer adoption rate across all client projects - 90% code completion accuracy vs. 50% with previous solutions - 50+ client projects reporting productivity improvements in less than 6 months
- DSO National Laboratories and Mistral / Mistral AI — complex, DSO needed to train its own multilingual large language models from scratch that could be deployed entirely on premise. To accelerate development, DSO partnered with Mistral to co-develop a specialized multilingual MoE model using Forge,
- Orange / Mistral AI — with 298 million customers worldwide. Orange uses Mistral models to generate personalized promotion messages that are individualized to recipients’ characteristics and profile and intended to promote sales activities. A tailor-made marketing campaign was
External links were grounded in one live web search. Search snippets are leads, not independently scraped page content; rejected links remain available in Review evidence.
Company, security, and trust
Ownership, policies, security signals, and public trust evidence.
| Signal | Finding | Source |
|---|---|---|
| Company and ownership | Co-founder and CEO. | source |
| Contact and support | Yes, the Mistral Enterprise plan includes custom SLAs, dedicated support, and private deployments. Contact us for more details. | source |
| Legal and privacy | Subject to fair usage limits and Mistral's Terms of Service. | source |
| Security and compliance | Customize every layer: models fine-tuned on your data, agents that sharpen with use, and full data residency in any cloud or datacenter. Your intelligence stays yours. | source |
| Uptime and status page | Oui, le plan Mistral Enterprise comprend des SLA personnalisés, une assistance dédiée et des déploiements privés.Contact us for more details. | source |
| Domain registered | 2019-05-15 | — |
| Domain expires | 2029-05-15 | — |
Every trust signal above is backed by a collected page or a registration record.
Documentation and product updates
Help resources, product guidance, and visible changes over time.
| Resource or update | Type | What is available | Date or freshness |
|---|---|---|---|
| API pricing / Mistral Docs | Documentation | Pricing for Mistral models through the API, organized by family. (source) | Not dated |
Not found in the collected public sources: public roadmap, changelog or release notes, status page, separate help center or knowledge base.
Technical
Technical and delivery signals
Delivery model, platforms, API, data ownership, reliability claims and release activity.
| Signal | Finding | Source |
|---|---|---|
| Delivery model (cloud, self-hosted, desktop, browser extension) | Available as cloud and on-premises deployment S7. | S7 |
| Platforms and apps | Not published on the collected pages. | — |
| API and extensibility (API, SDK, webhooks, MCP, plugins) | Offers API access and MCP compatibility S73 S103. | S73 |
| Data ownership and export | Enterprise deployments include full data residency S7. | S7 |
| Performance and reliability claims | Not published on the collected pages. | — |
| Release activity | Not published on the collected pages. | — |
Measured infrastructure
Domain, certificate, and hosting signals, measured directly.
| Signal | Finding |
|---|---|
| Domain registered | 2019-05-15 — Cloudflare, Inc |
| Domain expires | 2029-05-15 |
| First archived | 2022-03-24 — web archive |
| TLS certificate | Google Trust Services, valid, expires 2027-01-07 |
| Security headers | HSTS, CSP, X-Content-Type-Options, X-Frame-Options, Referrer-Policy |
| Server header | cloudflare |
| Mail handled by | mailstream-eu1.mxrecord.io, mailstream-central.mxrecord.mx |
| Nameservers | ivan.ns.cloudflare.com, ada.ns.cloudflare.com |
Not measurable: rdap: exceeded the overall budget; wayback: exceeded the overall budget; tls: exceeded the overall budget; dns: exceeded the overall budget; headers: exceeded the overall budget. A gap here means the lookup was unavailable, not that the finding is negative.
Technical site audit
A bounded multi-page crawl with page-level issues and URLs.
Scout — Bounded technical crawl
- Pages crawled: 15 of 15 maximum
- Technical health: 94/100
- Critical issues: 0
- Warnings: 11
- Notices: 18
Issues
| Severity | Issue | Value | Affected URL |
|---|---|---|---|
| warning | Page should have one primary H1 | 0 | https://mistral.ai/ |
| notice | Images are missing alt text | 44 | https://mistral.ai/ |
| notice | Meta description length should be reviewed | 31 | https://mistral.ai/contact/ |
| notice | Images are missing alt text | 4 | https://mistral.ai/contact/ |
| notice | Images are missing alt text | 57 | https://mistral.ai/research/ |
| notice | Meta description length should be reviewed | 31 | https://mistral.ai/contact/ |
| notice | Images are missing alt text | 4 | https://mistral.ai/contact/ |
| notice | Meta description length should be reviewed | 7 | https://mistral.ai/sovereign-ai/ |
| notice | Images are missing alt text | 17 | https://mistral.ai/sovereign-ai/ |
| notice | Images are missing alt text | 5 | https://mistral.ai/products/ |
| notice | Images are missing alt text | 26 | https://mistral.ai/our-method/ |
| notice | Images are missing alt text | 16 | https://mistral.ai/use-cases/ |
| notice | Images are missing alt text | 29 | https://mistral.ai/cloud/inference/ |
| notice | Images are missing alt text | 28 | https://mistral.ai/cloud/compute/ |
| notice | Images are missing alt text | 29 | https://mistral.ai/industry/energy/ |
| notice | Images are missing alt text | 26 | https://mistral.ai/industry/manufacturing/ |
| notice | Images are missing alt text | 21 | https://mistral.ai/industry/financial-services/ |
| notice | Images are missing alt text | 22 | https://mistral.ai/industry/public-sector/ |
| notice | Images are missing alt text | 4 | https://mistral.ai/customers/ |
| warning | Duplicate title in audit sample | 2 | https://mistral.ai/contact/ |
| warning | Duplicate title in audit sample | 2 | https://mistral.ai/contact/ |
| warning | Duplicate meta description in audit sample | 2 | https://mistral.ai/contact/ |
| warning | Duplicate meta description in audit sample | 2 | https://mistral.ai/contact/ |
| warning | Duplicate meta description in audit sample | 6 | https://mistral.ai/products/ |
| warning | Duplicate meta description in audit sample | 6 | https://mistral.ai/use-cases/ |
| warning | Duplicate meta description in audit sample | 6 | https://mistral.ai/industry/energy/ |
| warning | Duplicate meta description in audit sample | 6 | https://mistral.ai/industry/manufacturing/ |
| warning | Duplicate meta description in audit sample | 6 | https://mistral.ai/industry/financial-services/ |
| warning | Duplicate meta description in audit sample | 6 | https://mistral.ai/industry/public-sector/ |
Crawled pages
| Status | Title | H1s | Words | Response | URL |
|---|---|---|---|---|---|
| 200 | Mistral \ | #1 in Sovereign AI: Frontier Open Models You Own | 0 | 1053 | 115 ms |
| 200 | Contact us \ | Mistral | 1 | 560 | 232 ms |
| 200 | AI Research \ | Mistral | 1 | 1119 | 185 ms |
| 200 | Contact us \ | Mistral | 1 | 560 | 120 ms |
| 200 | Sovereign AI — Own the Intelligence You Build \ | Mistral | 1 | 1656 | 163 ms |
| 200 | Products \ | Mistral | 1 | 466 | 232 ms |
| 200 | Mistral services \ | Build your own AI, supported by Mistral. | 1 | 936 | 265 ms |
| 200 | Frontier AI LLMs, assistants, agents, services \ | Mistral AI | 1 | 526 | 360 ms |
| 200 | Inference \ | Mistral | 1 | 1956 | 252 ms |
| 200 | AI Cloud for training and inference at scale \ | Mistral | 1 | 993 | 325 ms |
| 200 | Energy \ | Mistral | 1 | 1006 | 219 ms |
| 200 | Manufacturing \ | Mistral | 1 | 1063 | 62 ms |
| 200 | Financial Services \ | Mistral | 1 | 975 | 185 ms |
| 200 | Public Sector \ | Mistral | 1 | 982 | 332 ms |
| 200 | Customer stories \ | Mistral | 1 | 1087 | 182 ms |
This is a bounded Scout crawl. Core Web Vitals and Lighthouse scores are not claimed unless a dedicated performance audit is connected.
Search Console and Analytics
First-party search and behavioral data connection state.
Modeled acquisition channels
| Channel | Estimated share |
|---|---|
| direct | 58.7% |
| search | 25.0% |
| search organic | 24.9% |
| search paid | 0.0% |
| social | 5.7% |
| social organic | 5.7% |
| social paid | 0.0% |
| referrals | 4.8% |
| paid referrals | 0.2% |
| 1.6% | |
| gen ai | 4.1% |
| affiliate | 0.0% |
| other | 0.0% |
First-party connections
| Integration | Status | Data available after connection |
|---|---|---|
| Google Search Console | Not connected | Clicks, impressions, CTR, average position, queries, pages, countries, and devices |
| Google Analytics | Not connected | Organic sessions, engagement, landing pages, conversions, geography, and devices |
No first-party Search Console or Analytics numbers are shown until the site owner authorizes the corresponding Google account. Scout traffic model values above are modeled estimates, not private analytics totals.
Limits and method
Limitations and open questions
Known constraints and missing evidence from the collected pages.
The collected pages do not mention any competitors by name. Evidence gaps include lack of customer reviews, detailed security certifications, and uptime information. Open questions remain about specific platform support and performance metrics.
Sources, method, and limitations
Evidence inventory, collection method, provider use, and constraints.
This report is based on 15 collected pages from mistral.ai, including product, pricing, and documentation pages. Static and browser rendering collectors were used; deep page fetch was disabled. External proof search was performed but no additional verification was completed. Limitations include incomplete rendering of some pages and lack of third-party review data.
- Official Websitehttps://mistral.ai/products/vibe/Verified
- Official Websitehttps://mistral.ai/about/Verified
- Official Websitehttps://mistral.ai/products/Verified
- Pricing Pagehttps://mistral.ai/pricing/Verified
- Pricing Pagehttps://mistral.ai/fr/pricing/Verified
- Pricing Pagehttps://mistral.ai/it/pricing/Verified
- Pricing Pagehttps://docs.mistral.ai/inference/pricingVerified
- Pricing Pagehttps://mistral.ai/fr/pricing/enterprise-deployments/Verified
- Pricing Pagehttps://mistral.ai/it/pricing/enterprise-deployments/Verified
- Official Websitehttps://mistral.ai/fr/solutions/Verified
- Official Websitehttps://mistral.ai/it/solutions/Verified
- Official Websitehttps://mistral.ai/solutions/coding/Verified
- Official Websitehttps://mistral.ai/solutions/document-ai/Verified
- Official Websitehttps://mistral.ai/solutions/speech/Verified
- Articlehttps://mistral.ai/customers/cisco/Verified
- Articlehttps://mistral.ai/customers/belfius/Verified
- Articlehttps://mistral.ai/customers/pretto/Verified
- Articlehttps://mistral.ai/customers/Verified
- Articlehttps://mistral.ai/customers/zalando/Verified
- Articlehttps://mistral.ai/customers/stellantis/Verified
- Articlehttps://mistral.ai/customers/ardian/Verified
- Articlehttps://mistral.ai/customers/capgemini/Verified
- Articlehttps://mistral.ai/customers/dso/Verified
- Articlehttps://mistral.ai/customers/orange/Verified
