Mistral AI vs OpenAI: Revenue, Profit and Business Model
A full-year revenue figure is unavailable here for Mistral AI. A full-year revenue figure is unavailable here for OpenAI.
Latest financial snapshot
Mistral AI
- Latest annual revenue
- Unavailable here
- Net income
- Unavailable here
- Net margin
- —
- Revenue growth
- —
OpenAI
- Latest annual revenue
- Unavailable here
- Net income
- Unavailable here
- Net margin
- —
- Revenue growth
- —
Financial summary
Mistral AI
Mistral is private and does not publish audited accounts. Reported figures: ARR of about $339 million (€300 million) in September 2025 (CNBC) and more than $400 million in January 2026, up from about $20 million a year earlier (CNBC Disruptor 50). CEO Arthur Mensch said in 2026 he expected ARR to exceed $1 billion by year-end. Funding history: ~$119 million (€105 million) seed (June 2023), ~$435 million (€385 million) Series A (December 2023), ~$678 million (€600 million) Series B at ~$6.55 billion (€5.8 billion) (June 2024), ~$1.92 billion (€1.7 billion) Series C led by ASML at ~$13.2 billion (€11.7 billion) (September 2025), $830 million of debt for data centres (March 2026) and a ~$3.39 billion (€3 billion) round led by Samsung at more than $23.7 billion (€21 billion) (September 2026).
OpenAI
OpenAI reports revenue as annualized run rate rather than audited GAAP figures. The company says it reached $1 billion of revenue within a year of launching ChatGPT, was generating about $1 billion per quarter by the end of 2024, and passed $20 billion of annualized revenue by the end of 2025. Axios reported the run rate was nearing $70 billion at the end of September 2026. Funding has scaled alongside: $110 billion announced in February 2026 at a $730 billion pre-money valuation grew to $122 billion committed at $852 billion post-money by March 31, 2026, anchored by Amazon, Nvidia and SoftBank. An employee tender of about $7 billion followed in August 2026 at the same valuation. The company remains unprofitable because compute spending outpaces revenue.
Revenue and profit by year
Mistral AI
| Year | Revenue | Net income | Margin | Growth | Source |
|---|---|---|---|---|---|
| FY2026 run-rate | $400M | — | — | — | Source |
| FY2025 run-rate | ~$339M | — | — | — | Source |
| FY2024 run-rate | $16M | — | — | — | Source |
OpenAI
| Year | Revenue | Net income | Margin | Growth | Source |
|---|---|---|---|---|---|
| FY2025 run-rate | $20B | — | — | — | Source |
| FY2024 run-rate | $4B | — | — | — | Source |
| FY2023 run-rate | $2B | — | — | — | Source |
Where the revenue comes from
Mistral AI
- Enterprise API Token Consumption (La Plateforme)~45%
Usage-based per-million-token API fees paid by software developers and enterprises querying Mistral Large, Mistral Small, and Codestral.
- Cloud Hyperscaler Royalties (Azure, AWS, Snowflake)~35%
Revenue-sharing royalties collected when enterprise clients deploy and consume Mistral models natively inside public cloud marketplaces.
- On-Premise Private Sovereign Licenses~20%
Annual recurring multi-million-dollar software licenses for defense, government, and banking customers deploying models in air-gapped environments.
OpenAI
- ChatGPT SubscriptionsMajor
Consumer, team, business, education, and enterprise subscriptions.
- API UsageMajor
Usage-based developer and enterprise model access through OpenAI APIs.
- Enterprise PlatformGrowing
Business deployments, governance, custom workflows, and platform services.
- Partnerships and LicensingEmerging
Strategic licensing, infrastructure partnerships, commerce, and outcome-based AI workflows.
Business model and strategy
Mistral AI
How it makes money
Mistral makes money in four ways. It charges per token for API access to its models on La Plateforme; it sells enterprise licences and custom deployments, often on-premise or in private clouds, to companies and governments that want control over their data; it earns through cloud marketplaces such as Microsoft Foundry, AWS Bedrock and Google Vertex AI;
Growth strategy
Mistral's growth strategy combines sovereign infrastructure with enterprise partnerships. It is building data centres in France and Sweden, expanded its Microsoft partnership in July 2026 through a multibillion-dollar deal under which Microsoft uses part of its European compute, signed Accenture as a deployment partner in February 2026, and is moving into industrial AI through the May 2026 acquisition of Emmi AI and…
Competitive advantage
Mistral's main advantages are its position as the leading non-US, non-Chinese frontier AI lab, which suits European governments and regulated industries; open-weight models that customers can run on their own hardware; a research team led by founders who worked on Chinchilla at DeepMind and LLaMA at Meta;
OpenAI
How it makes money
OpenAI earns revenue in three main ways. First, ChatGPT subscriptions: consumer tiers such as Plus and Pro, plus Business, Enterprise and Edu seats sold to organizations. Second, the API platform, where developers and companies pay per token to build OpenAI models (GPT, reasoning, image, audio and embeddings) into their own software.
Growth strategy
OpenAI is growing on four fronts: selling ChatGPT and Codex into enterprises, turning ChatGPT into a broader 'super app' with agents, shopping and apps inside the chat, building its own compute through the Stargate data center program and chip partnerships, and moving into hardware through the 2025 acquisition of Jony Ive's io.
Competitive advantage
OpenAI's advantage is distribution plus scale. ChatGPT is the most widely used AI assistant, with OpenAI citing more than 900 million weekly users in February 2026, which gives it a direct consumer channel that rivals have to buy or build.
Questions about Mistral AI vs OpenAI
How do Mistral AI SAS and OpenAI make money?
Mistral AI SAS and OpenAI generate revenue in fundamentally different ways. Mistral AI SAS: Mistral makes money in four ways. OpenAI: OpenAI earns revenue in three main ways.
Figures come from each company's filings and the sources linked beside them. Amounts reported in another currency are shown in US dollars at an approximate rate and marked with ~. Back to the Mistral AI vs OpenAI overview