AI API price index · model maker · read 10 Nov 2026
Mistral AI API pricing
19 prices as Mistral AI publishes them — 10 text models, 2 embeddings, 4 speech to text, 2 text to speech, 1 images and documents — read on its own pricing page on 10 Nov 2026. Its current text models cost $0.10 to $4.40 per million output tokens.
Free tier No free API tier with a dollar amount is stated. mistral.ai/pricing: 'Mistral's Free plan provides limited access to Vibe ... On the Free plan, you also get access to Mistral Studio.' mistral.ai/cloud/inference FAQ: 'Yes, in Studio, you can test-prompt general models in the agent playground, or try documents and speech in the dedicated OCR and audio playgrounds.' (Paid consumer plans on mistral.ai/pricing list monthly API credits, e.g. '$15 /mo in API credits'; the page layout did not make the plan-to-credit mapping unambiguous, so no figure is assigned here.)
What to know before you build on Mistral AI
- The API price table now lives on docs.mistral.ai; mistral.ai/pricing shows the consumer plans.
- Mistral Large 4 is $0.70 in, $0.07 cached and $2.10 out. The page no longer marks it as a sale or shows the old $1.36 and $4.18 list price (read 10 November).
- Mistral Large 3, Mistral Medium 3.5, Voxtral Mini Transcribe 2, Voxtral Mini Transcribe Realtime and Voxtral TTS were deprecated on 31 October 2026 and retire on 30 April 2027. They have left the price table, so the index no longer ranks them.
- Batch halves the price, cached input is 90% off, and the regional EU or US endpoints add 10%.
- Most of the lineup is open-weight, Mistral Large 4, Mistral Small 4 and the Ministral 3 models among them, so the same models can also be run on your own hardware.
- No free dollar allowance is stated; Studio's playgrounds are the way to test.
Against the other 31
Across the index, 390 current text models run from $0.03 to $50.00 per million output tokens — 1667×. Mistral AI’s cheapest, Ministral 3 3B at $0.10, has 5 cheaper text models elsewhere; its dearest, Z.ai GLM 5.3, costs $4.40. A cheaper model is a different model, not the same one for less — compare capability before price.
Speech to text across the index runs from $0.04 to $0.62 per audio hour; Mistral AI’s Voxtral Mini Transcribe 3 is $0.21.
Text to speech across the index runs from $11.00 to $100.00 per million characters; Mistral AI’s cheapest ranked voice, Voxtral TTS 3, is $13.00.
The same model elsewhere
Open-weight models Mistral AI sells that other hosts price too — the only like-for-like comparison in the index.
- GLM 5.3 — DeepInfra $0.90 / $4.00 · Cloudflare Workers AI $1.40 / $4.40 · Fireworks AI $1.40 / $4.40 · Mistral AI $1.40 / $4.40 · Nebius Token Factory $1.40 / $4.40 · Novita AI $1.40 / $4.40 · Perplexity $1.40 / $4.40 · Together AI $1.40 / $4.40 · Z.ai $1.40 / $4.40 · SambaNova Cloud $2.80 / $8.80 — 2.2× apart
Input / output per 1M tokens, standard tier.
Text models
| Model | Input | Cached input | Output | Note |
|---|---|---|---|---|
| Mistral Large 4 | $0.70 | $0.07 | $2.10 | — |
| Mistral Large 3legacy | $0.50 | $0.05 | $1.50 | Deprecated 31 Oct 2026, retires 30 Apr 2027, replaced by Mistral Large 4. Mistral no longer prints a price for it; this is the price last read on 16 Oct · Not ranked: legacy model, existing customers only. |
| Mistral Medium 3.5legacy | $1.50 | $0.15 | $7.50 | Deprecated 31 Oct 2026, retires 30 Apr 2027, replaced by Mistral Large 4. Mistral no longer prints a price for it; this is the price last read on 16 Oct · Not ranked: legacy model, existing customers only. |
| Mistral Small 4 | $0.15 | $0.015 | $0.60 | — |
| Ministral 3 14B | $0.20 | $0.02 | $0.20 | — |
| Ministral 3 8B | $0.15 | $0.015 | $0.15 | — |
| Ministral 3 3B | $0.10 | $0.01 | $0.10 | — |
| Z.ai GLM 5.3 | $1.40 | $0.14 | $4.40 | Third-party model hosted by Mistral |
| Codestral | $0.30 | $0.03 | $0.90 | — |
| Mistral Moderation 2 | $0.00 | $0.00 | $0.00 | Printed as Free · Not ranked: a safety classifier, not a general model. |
Embeddings
| Model | Input | Cached input | Note |
|---|---|---|---|
| Codestral Embed | $0.15 | $0.015 | — |
| Mistral Embed | $0.10 | $0.01 | — |
Speech to text
| Model | Price | Note |
|---|---|---|
| Voxtral Mini Transcribe 2legacy | $0.003 per minute | Per minute of audio input. Deprecated 31 Oct 2026, retires 30 Apr 2027, replaced by Voxtral Mini Transcribe 3. Mistral no longer prints a price for it; this is the price last read on 16 Oct · Not ranked: legacy model, existing customers only. |
| Voxtral Mini Transcribe 3 | $0.0035 per minute | Per minute of audio input; cached $0.00035 /Min; Public Preview |
| Voxtral Mini Transcribe Realtimelegacy | $0.006 per minute | Deprecated 31 Oct 2026, retires 30 Apr 2027, replaced by Voxtral Mini Transcribe Realtime 3. Mistral no longer prints a price for it; this is the price last read on 16 Oct · Not ranked: legacy model, existing customers only. |
| Voxtral Mini Transcribe Realtime 3 | $0.008 per minute | Per minute of audio input; cached $0.0008 /Min |
Text to speech
| Model | Price | Note |
|---|---|---|
| Voxtral TTSlegacy | $16.00 per 1M characters | Printed as input $0 /M Chars, output $16 /M Chars. Deprecated 31 Oct 2026, retires 30 Apr 2027, replaced by Voxtral TTS 3. Mistral no longer prints a price for it; this is the price last read on 16 Oct · Not ranked: legacy model, existing customers only. |
| Voxtral TTS 3 | $13.00 per 1M characters | Printed as input $13 /M Chars, cached $1.3 /M Chars |
Images and documents
| Model | Price | Note |
|---|---|---|
| OCR 4.1 | $4.00 per 1K pages | Per 1,000 pages; cached $0.4 /1000 Pages. mistral.ai/cloud/inference also prints 'Document AI $5 / 1000 pages' |
12 of these 19 prices are ranked in the index; the rest say why in their note. Text and embeddings per 1M tokens unless the row says otherwise.
Discounts and surcharges, as published
- Batch: 'High-volume processing, at half price' / 'Batch processing, for high-volume work, cuts the price by 50%' (the docs pricing page also has a Batch toggle).
- Cached input: 'cutting input costs by 90%' — cached-input prices are printed per model in the rows.
- Mistral Large 4 is on a 'Temporary sale price. The struck-through amount is the original price.' (tooltip on the docs pricing page; no end date given).
Where these numbers come from
Read on docs.mistral.ai/inference/pricing on 10 Nov 2026, checked against 5 more of Mistral AI’s own pages. Macrostack measures nothing and is paid by no provider on this index. Prices change; the date is the claim.
The other providers in the index
- The AI API price index — every provider, cheapest first, and the download
- Free LLM APIs — which providers let you use a model free, and the catch with each
- OpenAI API pricing — model maker
- Anthropic API pricing — model maker
- Google Gemini API pricing — model maker
- Cohere API pricing — model maker
- Groq API pricing — open-model host
- Together AI API pricing — open-model host
- Fireworks AI API pricing — open-model host
- Replicate API pricing — open-model host
- Deepgram API pricing — voice
- ElevenLabs API pricing — voice
- DeepSeek API pricing — model maker
- xAI API pricing — model maker
- Azure OpenAI API pricing — cloud platform
- Amazon Bedrock API pricing — cloud platform
- Voyage AI API pricing — model maker
- AssemblyAI API pricing — voice
- Speechmatics API pricing — voice
- Cartesia API pricing — voice
- MiniMax API pricing — model maker
- Z.ai API pricing — model maker
- Cerebras API pricing — open-model host
- Novita AI API pricing — open-model host
- Moonshot AI (Kimi) API pricing — model maker
- Alibaba Cloud Model Studio (Qwen) API pricing — model maker
- Perplexity API pricing — model maker
- DeepInfra API pricing — open-model host
- Cloudflare Workers AI API pricing — open-model host
- Nebius Token Factory API pricing — open-model host
- SambaNova Cloud API pricing — open-model host
- Google Vertex AI (Agent Platform) API pricing — cloud platform
- Azure AI Foundry Models API pricing — cloud platform
Questions
- How much does the Mistral AI API cost?
- On 10 Nov 2026, Mistral AI's current text models cost $0.10 to $4.40 per million output tokens at the standard tier (Ministral 3 3B to Z.ai GLM 5.3). The full list, with input and cached-input prices, is on this page.
- Does Mistral AI have a free tier?
- What Mistral AI's own pages say: No free API tier with a dollar amount is stated. mistral.ai/pricing: 'Mistral's Free plan provides limited access to Vibe ... On the Free plan, you also get access to Mistral Studio.' mistral.ai/cloud/inference FAQ: 'Yes, in Studio, you can test-prompt general models in the agent playground, or try documents and speech in the dedicated OCR and audio playgrounds.' (Paid consumer plans on mistral.ai/pricing list monthly API credits, e.g. '$15 /mo in API credits'; the page layout did not make the plan-to-credit mapping unambiguous, so no figure is assigned here.)
- Does Mistral AI offer discounts?
- Batch: 'High-volume processing, at half price' / 'Batch processing, for high-volume work, cuts the price by 50%' (the docs pricing page also has a Batch toggle). Cached input: 'cutting input costs by 90%' — cached-input prices are printed per model in the rows. Mistral Large 4 is on a 'Temporary sale price. The struck-through amount is the original price.' (tooltip on the docs pricing page; no end date given).