BroadRouter

Mistral AI API pricing

Updated

Mistral AI lists standard per-token prices on its own pricing page for 9 models. The cheapest is Ministral 3 3B at USD 0.10 input and USD 0.10 output per 1M tokens; the dearest is GLM 5.3 at USD 1.40 / USD 4.40. It publishes cached-input prices for all of them, plus batch and priority tiers (no flex). Checked 2026-11-15.

Mistral AI is a model maker: 6 of the 9 models here are its own. By maker: Mistral AI 6, Z.ai 2, DeepSeek 1. 27 prices, each read from Mistral AI's own pricing page, checked 2026-11-15. BroadRouter is BroadNet's own site and does not sell Mistral AI's services. Automated re-check 2026-11-16: 27 of 27 prices still matched the provider's page (8 of them on pages that load prices with JavaScript, re-read with a page renderer within the last 7 days).

Models with a standard price
9
Median input / output
USD 0.20 / USD 0.60
Cached-input prices
9 of 9
Cheapest host for
3 of 6 shared models

Which models is Mistral AI the cheapest provider for?

Of the 6 models that Mistral AI and at least one other provider both price at the standard tier, Mistral AI has the lowest price for 3, ranked on a blend of 3 input tokens to 1 output token (all of them tied with another provider at the same price; the maker's own API is listed first on a tie). It is the dearest for 1. Another 3 models have a standard price from Mistral AI only.

Models where Mistral AI is the cheapest provider
ModelMistral AI input / outputNext cheapestSaving on 3:1 blendProviders
Ministral 3 14BUSD 0.20 / USD 0.20Amazon Bedrock USD 0.20 / USD 0.20tie3
Ministral 3 3BUSD 0.10 / USD 0.10Amazon Bedrock USD 0.10 / USD 0.10tie2
Ministral 3 8BUSD 0.15 / USD 0.15Amazon Bedrock USD 0.15 / USD 0.15tie2

The 6 models Mistral AI sells that the most other providers also sell, with Mistral AI's rank among them on the 3:1 blend.

Mistral AI on the most widely hosted models
ModelMistral AIRankCheapest
GLM 5.3USD 1.40 / USD 4.406 of 24DeepInfra USD 0.5625 / USD 2.50
GLM 5.3 FlashUSD 0.15 / USD 0.503 of 20GMI Cloud USD 0.075 / USD 0.25
DeepSeek V4.1 Flash 1017USD 0.30 / USD 1.206 of 16DeepInfra USD 0.098 / USD 0.294
Ministral 3 14BUSD 0.20 / USD 0.201 of 3Mistral AI (this page)
Ministral 3 3BUSD 0.10 / USD 0.101 of 2Mistral AI (this page)
Ministral 3 8BUSD 0.15 / USD 0.151 of 2Mistral AI (this page)

Which model makers does Mistral AI sell?

Mistral AI prices models from 3 makers. Mistral AI leads with 6 models. Standard prices, USD per 1M input / output tokens.

Mistral AI models by maker
MakerModelsCheapest modelMedian input / output
Mistral AI6Ministral 3 3B USD 0.10 / USD 0.10USD 0.175 / USD 0.40
Z.ai2GLM 5.3 Flash USD 0.15 / USD 0.50USD 0.775 / USD 2.45
DeepSeek1DeepSeek V4.1 Flash 1017 USD 0.30 / USD 1.20—

Mistral AI price list

USD per 1M tokens: 27 prices for 9 models, one row per model with any other service tiers stacked under the standard price. Sources checked 2026-11-15 unless a row says otherwise. "vs maker" compares a standard price with the model maker's own API on the 3:1 blend; a minus means Mistral AI is cheaper. Of 3 hosted model prices with a maker price to compare, 0 are below the maker's and 0 above.

Mistral AI API prices by model
ModelTierInputCachedOutputContextvs makerSource
Codestral 2508standard
batch
priority
USD 0.30
USD 0.15
USD 0.525
USD 0.03
USD 0.015
USD 0.0525
USD 0.90
USD 0.45
USD 1.575
131,072own modelsource 2026-11-14, 2026-11-15
DeepSeek V4.1 Flash 1017standard
batch
priority
USD 0.30
USD 0.15
USD 0.525
USD 0.007
USD 0.0035
USD 0.01225
USD 1.20
USD 0.60
USD 2.10
1,048,576samesource 2026-11-14, 2026-11-15
GLM 5.3standard
batch
priority
USD 1.40
USD 0.70
USD 2.45
USD 0.14
USD 0.07
USD 0.245
USD 4.40
USD 2.20
USD 7.70
1,048,576samesource 2026-11-14, 2026-11-15
GLM 5.3 Flashstandard
batch
priority
USD 0.15
USD 0.075
USD 0.2625
USD 0.03
USD 0.015
USD 0.0525
USD 0.50
USD 0.25
USD 0.875
1,048,576samesource 2026-11-14, 2026-11-15
Ministral 3 14Bstandard
batch
priority
USD 0.20
USD 0.10
USD 0.35
USD 0.02
USD 0.01
USD 0.035
USD 0.20
USD 0.10
USD 0.35
262,144own modelsource 2026-11-14, 2026-11-15
Ministral 3 3Bstandard
batch
priority
USD 0.10
USD 0.05
USD 0.175
USD 0.01
USD 0.005
USD 0.0175
USD 0.10
USD 0.05
USD 0.175
262,144own modelsource 2026-11-14, 2026-11-15
Ministral 3 8Bstandard
batch
priority
USD 0.15
USD 0.075
USD 0.2625
USD 0.015
USD 0.0075
USD 0.02625
USD 0.15
USD 0.075
USD 0.2625
262,144own modelsource 2026-11-14, 2026-11-15
Mistral Large 4standard
batch
priority
USD 0.70
USD 0.35
USD 1.225
USD 0.07
USD 0.035
USD 0.1225
USD 2.10
USD 1.05
USD 3.675
1,048,576own modelsource 2026-11-14, 2026-11-15
Mistral Small 4standard
batch
priority
USD 0.15
USD 0.075
USD 0.2625
USD 0.015
USD 0.0075
USD 0.02625
USD 0.60
USD 0.30
USD 1.05
262,144own modelsource 2026-11-14, 2026-11-15
  • Cached "—" means Mistral AI's page shows no cache price for that row; budget cached tokens at the input price.

Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).

What service tiers does Mistral AI publish?

  • batch (9 rows). Across 9 models with both prices, the median batch price is 50% below standard. Mistral AI row note: “Context from the docs model card.”
  • priority (9 rows). Across 9 models with both prices, the median priority price is 1.75× standard. Mistral AI row note: “Context from the docs model card.”

Cached input: 9 of 9 standard prices include a cache-read price, a median 10% of the input price. What cached input means.

What we recorded about Mistral AI's pricing page: Prices from the Mistral docs API pricing page (Standard, Batch and Priority tabs, USD), rendered with Playwright; mistral.ai/pricing no longer shows API prices. Columns are Input, Cached input, Output per 1M tokens. Mistral also hosts third-party Z.ai and DeepSeek models; those rows carry the original maker. Deprecated models with no price on the page (Mistral Large 3, Mistral Medium 3.5, Leanstral 1.5) are left out.

Is Mistral AI cheaper than the model makers' own APIs?

Mostly it charges the same as the maker. Of 3 standard prices for models whose maker sells its own API, 0 are below the maker's price, 0 above and 3 the same, on the 3:1 blend. The median difference is zero.

BroadNet Inference and Mistral AI

BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference also serves 1 model that Mistral AI prices: GLM 5.3 at USD 0.78 / USD 3.46 against Mistral AI's USD 1.40 / USD 4.40 (BroadNet is cheaper on the 3:1 blend). BroadNet prices are from its rate card, synced 2026-11-14. BroadNet Inference pricing.

More on Mistral AI

Quick answers

Is Mistral AI OpenAI-compatible?

Not stated. Mistral's API reference documents POST /v1/chat/completions but does not call itself OpenAI-compatible, so test before relying on it (Mistral AI API reference, checked 2026-11-15).

What is Mistral AI's API base URL?

https://api.mistral.ai/v1, as Mistral AI documents it (Mistral AI API reference, checked 2026-11-15).

Does Mistral AI keep API prompts and outputs?

Yes, for 30 days. Except for specific APIs, Mistral keeps API 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)". Agents API data is kept until you close your account. Zero data retention is available on paid plans for supported stateless calls, after review (Mistral AI Privacy Policy, checked 2026-11-14).

Does Mistral AI train on API data?

It can. Mistral says it may use free-mode input and output to train its models, with an opt-out. For pay-as-you-go it says you can opt out at any time but does not state the default; the opt-out is the "Anonymous improvement data" toggle under Privacy in the Admin panel (Mistral Help Center, training on user data, checked 2026-11-14).

Does Mistral AI have a free tier?

Free mode includes a monthly usage budget for trying Vibe and Studio, but "It doesn't include API access." Creating API keys requires billing activation and pay as you go. Limits are shown in the Admin Panel (API › Limits), not published as numbers on the page (Mistral usage limits docs, checked 2026-11-14).

Questions

How much does the Mistral AI API cost per token?

Mistral AI's standard prices run from USD 0.10 input and USD 0.10 output per 1M tokens (Ministral 3 3B) to USD 1.40 / USD 4.40 (GLM 5.3). The median across 9 standard prices is USD 0.20 input and USD 0.60 output per 1M, which is USD 0.0000002 per input token. Input, cached input and output are billed separately.

What is the cheapest model on Mistral AI?

Ministral 3 3B, at USD 0.10 input and USD 0.10 output per 1M tokens (USD 0.10 on a 3:1 input-to-output blend). The cheapest from other makers: GLM 5.3 Flash (Z.ai) at USD 0.15 / USD 0.50; DeepSeek V4.1 Flash 1017 (DeepSeek) at USD 0.30 / USD 1.20.

Is Mistral AI cheaper than other providers?

It depends on the model. Of 6 models that other providers also price, Mistral AI is the cheapest standard price for 3 (all of them tied with another provider), for example Ministral 3 14B, Ministral 3 3B and Ministral 3 8B; another provider is cheaper for 3; and it is the dearest for 1. Its median rank is 2 of a median 9.5 providers, on a 3:1 input-to-output blend.

Does Mistral AI have batch, cached or priority pricing?

Cached input: yes, for 9 of 9 standard prices, at a median 10% of the input price. Other tiers: batch on 9 rows (median input 50% below standard); priority on 9 rows (median input 1.75× standard).

How much does a chatbot cost per month on Mistral AI?

For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 5.94 on Ministral 3 3B, USD 17.01 on Mistral Small 4 (the middle of Mistral AI's list) and USD 137.16 on GLM 5.3. Try your own numbers.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity