What exactly is interleaved sliding-window attention?

It alternates between layers that use full attention (every token sees every other token) and layers that use windowed attention (each token only sees nearby tokens). This combination preserves long-range reasoning while dramatically reducing memory consumption.

How does Ministral 8B compare to standard 8B transformers on inference speed?

Mistral designed the sliding-window pattern specifically to outperform standard architectures on speed and memory at this scale. The exact advantage depends on hardware and serving stack, but the architectural benefit is most pronounced on long sequences.

What licensing does Ministral 8B offer that the 3B does not?

Ministral 8B includes both the Mistral Commercial License and the Mistral Research License. The 3B variant's licensing is more limited.

When should I choose Ministral 8B over scaling up to a larger model like Mistral Small?

Choose Ministral 8B when you need the memory efficiency that sliding-window attention provides. For workloads that benefit from more capability, a larger model may be more cost-effective per unit of capability.

Mistral

Ministral 8B

Ministral 8B brings an interleaved sliding-window attention architecture to edge inference, delivering faster and more memory-efficient processing across its full context window of 128K tokens at $0.15 per million tokens.

View API reference

Input and output price: Input $0.15, Output $0.15, Per 1M tokens
24h uptime: Loading AI Gateway uptime

Get API key Read the docs

import { streamText } from 'ai'

const result = streamText({
  model: 'mistral/ministral-8b',
  prompt: 'Why is the sky blue?'
})

Read docs

Overview Playground Providers Uptime Status Throughput Latency API Similar About FAQ

Copy link to headingPlayground

Try out Ministral 8B by Mistral. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.

Ministral 8B

Copy link to headingProviders

Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.

Provider

Context	Max Output	Latency	Throughput	Input	Output	Cache	Web Search	Capabilities	ZDR	No Training	Free Tier	Release Date

Mistral

Legal:Terms•Privacy

128K

0.3 s

97 tps

$0.15/M

—

10/16/2024

Getting started

Call Ministral 8B through AI Gateway with the AI SDK generateText and streamText functions, or through the OpenAI Chat Completions, OpenAI Responses, and Anthropic Messages APIs by changing the base URL. AI Gateway authenticates the request and routes it to an available provider.

Install the AI SDK (pnpm add ai dotenv), create an API key from the API Keys page, and set it as AI_GATEWAY_API_KEY in your environment. Full setup is covered in the text generation quickstart.

index.ts

import { generateText } from 'ai';
import 'dotenv/config';

async function main() {
  const result = await generateText({
    model: 'mistral/ministral-8b',
    prompt: 'Why is the sky blue?',
  });

  console.log(result.text);
}

main().catch(console.error);

Top-level parameters

The same Ministral 8B request in each API format AI Gateway supports.

top-level-params.ts

import { generateText } from 'ai';
import 'dotenv/config';

async function main() {
  const result = await generateText({
    model: 'mistral/ministral-8b',
    system: 'You are a concise technical assistant.',
    prompt: 'Summarize the tradeoffs between static generation and SSR.',
    maxOutputTokens: 1024,
    temperature: 0.5,
  });

  console.log(result.text);
}

main().catch(console.error);

Standard parameters like prompt, messages, temperature, and tools work as documented in the AI SDK docs. These are the parameters with model-specific behavior.

Parameter	Type	Required	Description
`model`	`string`	Yes	Model ID in the form `creator/model`, e.g. `mistral/ministral-8b`. AI Gateway routes the request to an available provider.
`maxOutputTokens`	`number`	No	Hard cap on generated tokens. Ministral 8B supports up to `4,000` output tokens.
`providerOptions`	`Record<string, JSONValue>`	No	AI Gateway routing options under `gateway`, plus any provider-native options under the provider’s own namespace — see the table below.

Input limits

Input	Formats	Sources	Max count	Max size	Limits
Text	—	—	—	—	Prompt and response share the 128K-token context window
Image	—	URL, base64, Uint8Array	—	—	Sent as image parts in messages; counts as input tokens

Provider options

Set AI Gateway routing options under providerOptions.gateway. For provider-specific options, pass them under the provider’s namespace as documented by the AI SDK.

Learn more in the AI SDK mistral provider docs.

provider-options.ts

import { generateText } from 'ai';
import 'dotenv/config';

async function main() {
  const result = await generateText({
    model: 'mistral/ministral-8b',
    prompt: 'Why is the sky blue?',
    providerOptions: {
      gateway: {
        only: ['mistral'],
      },
    },
  });

  console.log(result.text);
}

main().catch(console.error);

These AI Gateway routing options apply to every model. Provider-specific options pass through under the provider’s own namespace (for example providerOptions.anthropic) exactly as documented by the AI SDK.

Parameter	Type	Required	Description
`providerOptions.gateway.only`	`string[]`	No	Restrict routing to these provider slugs. Requests fail over only within the listed providers.
`providerOptions.gateway.order`	`string[]`	No	Preferred provider order. Listed providers are tried first; unlisted providers remain available as fallbacks.
`providerOptions.gateway.sort`	`'cost' \| 'ttft' \| 'tps'`	No	Rank candidate providers by price, time to first token, or tokens per second instead of the default routing order.
`providerOptions.gateway.zeroDataRetention`	`boolean`	No	Route only to providers with a zero-data-retention policy for this model.

Routing across providers

AI Gateway serves the same model through multiple providers and fails over automatically. order expresses a preference while keeping every provider eligible; only is a hard allowlist — if none of the listed providers are available the request fails instead of falling back.

Options under a provider's own namespace (for example providerOptions.anthropic) are forwarded to that provider with the request. Providers ignore option namespaces that don't apply to them, so it is safe to set provider options alongside gateway routing options.

Image input

Send images alongside text as message parts. Images count as input tokens.

image-input.ts

import { generateText } from 'ai';
import 'dotenv/config';

async function main() {
  const result = await generateText({
    model: 'mistral/ministral-8b',
    messages: [
      {
        role: 'user',
        content: [
          { type: 'text', text: 'Describe this image.' },
          { type: 'image', image: 'https://example.com/photo.jpg' },
        ],
      },
    ],
  });

  console.log(result.text);
}

main().catch(console.error);

Tool calling

Expose tools the model can call. Define each tool’s inputs with a Zod schema.

tool-calling.ts

import { generateText, tool } from 'ai';
import { z } from 'zod';
import 'dotenv/config';

async function main() {
  const result = await generateText({
    model: 'mistral/ministral-8b',
    prompt: 'What is the weather in San Francisco?',
    tools: {
      getWeather: tool({
        description: 'Get the current weather for a location',
        inputSchema: z.object({ location: z.string() }),
        execute: async ({ location }) => ({ location, temperatureC: 18 }),
      }),
    },
  });

  console.log(result.text);
}

main().catch(console.error);

Copy link to headingMore models by Mistral

Model

Context	Latency	Throughput	Input	Output	Cache	Web Search	Capabilities	Providers	ZDR	No Training	Free Tier	Release Date

mistral/mistral-large-3

256K

0.5 s

80 tps

$0.50/M

$1.50/M

—

12/02/2025

mistral/ministral-14b

256K

0.3 s

113 tps

$0.20/M

—

12/02/2025

mistral/codestral-embed

—

$0.15/M

—

05/28/2025

mistral/ministral-3b

128K

0.3 s

253 tps

$0.10/M

—

10/16/2024

mistral/mistral-small

32K

0.3 s

28 tps

$0.10/M

$0.30/M

—

09/17/2024

mistral/mistral-embed

—

$0.10/M

—

12/11/2023

Copy link to headingAbout Ministral 8B

Released October 16, 2024, Ministral 8B sits between the 3B and 14B models in Mistral's edge lineup. What sets Ministral 8B apart is its architecture: an interleaved sliding-window attention mechanism engineered for inference speed and memory efficiency.

Standard full-attention transformers require every token to attend to every other token, scaling quadratically with sequence length. Sliding-window attention limits each token's attention span, cutting memory usage. The interleaved design alternates between full-attention and windowed layers, preserving the ability to reason over long-range dependencies while keeping the memory footprint practical.

Ministral 8B uses its full context window of 128K tokens and supports function calling, knowledge retrieval, and commonsense reasoning.

Ministral 8B carries dual licensing: the Mistral Commercial License for production and the Mistral Research License for non-commercial work. This offers more flexibility than the 3B variant.

Copy link to headingWhat To Consider When Choosing a Provider

Configuration: For workloads processing long documents or extended conversation histories, Ministral 8B's sliding-window architecture reduces the memory pressure typical of long-context inference.
Zero Data Retention: Zero Data Retention is available for this model. It is offered on a per-provider and model basis. See the documentation for details.
Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.

Copy link to headingWhen to Use Ministral 8B

Best for

Long-context processing: Sliding-window attention keeps memory footprint manageable when processing long inputs
Deeper reasoning than 3B: Tasks requiring more depth than Ministral 3B can provide
Function calling and tool use: With better accuracy than the 3B variant
Dual licensing research use cases: Covered by the Commercial and Research licenses

Consider alternatives when

Smallest footprint and lowest cost: You need the absolute minimum (consider Ministral 3B)
Image understanding: Vision is required (consider Ministral 14B)

Copy link to headingConclusion

Ministral 8B earns its place through architectural innovation rather than just parameter scaling. The sliding-window attention design makes long-context inference more memory-efficient than standard transformers at this size.

Copy link to headingFrequently Asked Questions

What exactly is interleaved sliding-window attention?
It alternates between layers that use full attention (every token sees every other token) and layers that use windowed attention (each token only sees nearby tokens). This combination preserves long-range reasoning while dramatically reducing memory consumption.
How does Ministral 8B compare to standard 8B transformers on inference speed?
Mistral designed the sliding-window pattern specifically to outperform standard architectures on speed and memory at this scale. The exact advantage depends on hardware and serving stack, but the architectural benefit is most pronounced on long sequences.
What licensing does Ministral 8B offer that the 3B does not?
Ministral 8B includes both the Mistral Commercial License and the Mistral Research License. The 3B variant's licensing is more limited.
What is the context window of Ministral 8B?
128K tokens.
When should I choose Ministral 8B over scaling up to a larger model like Mistral Small?
Choose Ministral 8B when you need the memory efficiency that sliding-window attention provides. For workloads that benefit from more capability, a larger model may be more cost-effective per unit of capability.

Your use is subject to Mistral's Terms & Privacy Policies.

Agent Stack

Core Platform

Tools

Learn

Build

Explore

Ministral 8B

Copy link to headingPlayground

Copy link to headingProviders

Copy link to headingUptime24 hours

Copy link to headingThroughput24 hours

Copy link to headingLatency24 hours

Getting started

Top-level parameters

Input limits

Provider options

Routing across providers

Image input

Tool calling

Copy link to headingMore models by Mistral

Copy link to headingAbout Ministral 8B

Copy link to headingWhat To Consider When Choosing a Provider

Copy link to headingWhen to Use Ministral 8B

Best for

Consider alternatives when

Copy link to headingConclusion

Copy link to headingFrequently Asked Questions