Skip to content

Mistral AI

Pricing page

218 listings across 81 models

Data policies

from the provider's published policies

Headquarters: FRPrivacy policyTerms of serviceStatus page

On training

We’ve introduced a user control which allows you to object to the use of your input and output data for model training directly from your account.

Possible sources (2)
On keeping your data

Except for specific APIs, we keep your Input and Output for the period necessary to generate the Output and then for thirty (30) rolling days to monitor abuse (unless zero data retention is activated).

Source: Mistral AI privacy policy

More from Mistral AI on keeping your data (3)

we keep your Input and Output for the period necessary to generate the Output and then for thirty (30) rolling days to monitor abuse (unless zero data retention is activated).

Possible sources (2)

we keep your Input and Output for the period necessary to generate the Output and then for thirty (30) rolling days to monitor abuse

Possible sources (2)

the Personal Data will no longer be accessible upon the expiry of a thirty (30) days period following the termination

Possible sources (2)
Other statements

Mistral complies with both SOC 2 Type II and ISO 27001/27701 frameworks.

Possible sources (2)
More statements from Mistral AI (1)

Mistral AI will maintain an up-to-date list of its Subprocessors on Mistral AI’s website

Possible sources (2)

We take these statements from Mistral AI's own published policies each time we refresh the data, so they can change or disappear when those pages do. Check with the provider before you rely on any of this.

Page 1 of 4. The lists below split this page's listings into those sold by Mistral AI, run by Mistral AI, or sent to Mistral AI by a marketplace.

Sold by Mistral AI

34 listings on this page sold by Mistral AI.

Default order. Select a score for details.

About these columns
llminfo IQ
llminfo IQ (capability index). How llminfo IQ is computed
Coding IQ
llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
Value
Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
Yap
Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
$0.30cache read $0.007$1.201M
GLM 5.3mistral/zai-glm-5-3
$1.40cache read $0.14$4.401M
GLM 5.2
$1.40cache read $0.14$4.401M
GLM 5.3 Flashmistral/zai-glm-5-3-flash
$0.15cache read $0.03$0.501M
GLM 5mistral/zai-glm-5
$1.40cache read $0.14$4.401M
$0.70cache read $0.07$2.101M
Devstral 2 123b Instruct 2512devstral-2512
$0.40cache read $0.04$2.00262KNot measuredNot measured
Mistral Medium 3.1mistral-medium-2508
Not enough data yet$0.40$2.00262KNot measured
Mistral Large 3mistral-large-2512
$0.50cache read $0.05$1.50262KNot measured
Not enough data yet——Not published by the providerNot measuredNot measured
Codestral 25.01codestral
Not enough data yet$0.30$0.90Not published by the providerNot measured
Mistral Large 3
Not enough data yet$0.50cache read $0.05$1.50262KNot measuredNot measured
Mistral Medium 3
Not enough data yet$0.40cache read $0.15$2.00262KNot measured
Ministral 14Bministral-14b-2512 · 2512
Not enough data yet$0.20cache read $0.02$0.20262KNot enough data yetNot measuredNot measured
$1.50cache read $0.15$7.50262KNot measured
$2.39cache read $0.15$7.17262KNot measured
Not enough data yet$2.00$6.00Not published by the providerNot measuredNot measured
Not enough data yet$100.00$300.00Not published by the providerNot enough data yetNot measuredNot measured
Not enough data yet$0.075$0.20Not published by the providerNot measuredNot measured
Mistral Medium 3mistral-medium-2505
Not enough data yet$0.40$2.00131KNot measuredNot measured
Devstral Small 2505devstral-small
$0.06$0.12Not published by the providerNot measuredNot measured
$2.00$5.00Not published by the providerNot measuredNot measured
Magistral Small 2509magistral-small
Not enough data yet$0.50$1.50Not published by the providerNot measuredNot measured
Not enough data yet$0.70$0.70Not published by the providerNot enough data yetNot measuredNot measured
Show all 34 listings

Showing 24 of 34: one listing for each model.

Run by Mistral AI

66 listings on this page where Mistral AI runs the model. Another provider sells these listings.

Default order. Select a score for details.

About these columns
llminfo IQ
llminfo IQ (capability index). How llminfo IQ is computed
Coding IQ
llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
Value
Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
Yap
Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
 Hugging Face$0.30$1.201M
GLM 5.3
 Cortecs AI——1M
GLM 5.2
 Hugging Face$1.40$4.401M
 Cortecs AI——1M
Mistral Large 4mistral-large-4-0
 Cortecs AI——1M
Mistral Medium 3.1eu
 OpenRouterNot enough data yet$0.44cache read $0.044$2.20131KNot enough data yetNot measuredNot measured
Devstral 2 123b Instruct 2512mistral/devstral-2512
 Eden AI$0.40cache read $0.04$2.00262KNot measuredNot measured
Mistral Large 3mistral/mistral-large-2512
 Eden AI$0.50cache read $0.05$1.50262KNot measured
Devstral Mediummistral/devstral-medium-latest@eu · latest · 🇪🇺 eu
 Eden AINot enough data yet$0.44$2.20262KNot measuredNot measured
Codestral 25.01codestral-2508 · 2508
 Cortecs AINot enough data yet——256KNot measured
Mistral Medium 3
 Eden AINot enough data yet$0.44cache read $0.044$2.20262KNot measured
Ministral 14Bministral-14b-2512 · 2512
 Cortecs AINot enough data yet——256KNot enough data yetNot measuredNot measured
 Cortecs AI——256KNot measured
Mistral Mediummistral/mistral-medium-2604 · 2604
 Eden AI$1.50cache read $0.15$7.50262KNot measured
Devstral Small 2505mistral/devstral-small-latest · latest
 Requesty$0.11$0.33131KNot measuredNot measured
Magistral Mediummistral/magistral-medium-latest@eu · latest · 🇪🇺 eu
 Eden AI$1.65cache read $0.165$8.25262KNot measuredNot measured
Show all 66 listings

Showing 16 of 66: one listing for each model.

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)

Around the web

  • Mistral: Frontier AI LLMs, assistants, agents, services
    The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, …The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with open models.

    google.com

  • Mistral AI
    Mistral AI SAS (French: [mistʁal] ) is a French artificial intelligence (AI) company headquartered in …Mistral AI SAS (French: [mistʁal] ) is a French artificial intelligence (AI) company headquartered in Paris. Founded in 2023, it develops large language models ...Read more

    google.com

  • Mistral AI
    Mistral AI · 712 903 abonnés sur LinkedIn. Nous développons les meilleurs modèles d'intelligence artificielle …Mistral AI · 712 903 abonnés sur LinkedIn. Nous développons les meilleurs modèles d'intelligence artificielle générative.

    google.com

  • Mistral AI (@MistralAI) / X
    Meet Mistral AI at NVIDIA GTC 2026 to learn about our end-to-end AI solutions, discover …Meet Mistral AI at NVIDIA GTC 2026 to learn about our end-to-end AI solutions, discover how we can design and deploy AI to fit your specific needs, and see ...

    google.com

  • mistralai (Mistral AI_)
    Our first flaship models handling instruction-following, reasoning, and coding in a single set of opened-weights. …Our first flaship models handling instruction-following, reasoning, and coding in a single set of opened-weights. mistralai/Mistral-Medium-3.5- ...Read more

    google.com

How this data works