# Mistral Large 4

Mistral Large 4 is Mistral's open-weight, general-purpose multimodal model, a granular mixture-of-experts with 52B active and 1.05T total parameters plus a 1.6B vision encoder. It takes text and image input over a long context, can reason before answering or answer directly, and supports function calling and structured outputs. Its published context window is 524,288 tokens. The advertised input rate is $0.68 per million input tokens. The advertised output rate is $2.09 per million output tokens.

- Model identifier: ` mistral/mistral-large-4 `
- Creator: Mistral
- Providers: Mistral
- Model type: Text generation
- Canonical model page: [Mistral Large 4](<https://impossibl.com/mistral/mistral-large-4>)
- API availability: Listed as serving through Impossibl.
- Creator release date: 2026-10-06

## Providers

Mistral hosts Mistral Large 4. Impossibl routes your request intelligently and handles failover automatically when a provider is down.

| Provider | Endpoint location | Routes | Verified context window | Tool calling |
| --- | --- | --- | --- | --- |
| Mistral | Not specified | 1 | Not verified | Not verified |

The model price applies whichever provider serves the request. Endpoint locations do not guarantee inference residency.

## Specs and Capabilities

- Context window: 524,288 tokens
- Maximum output tokens: Not published in the public catalog; this is separate from the context window.
- Inputs: text, image
- Outputs: text
- Tool calling: Not published

## API access

Use ` mistral/mistral-large-4 ` as the model identifier through the Impossibl API.

Published API endpoints:

- ` /v1/chat/completions `

[API quickstart](<https://impossibl.com/docs/agent-quickstart>) · [Models and billing documentation](<https://impossibl.com/docs/models>)

## Published API prices

All rates are in USD. An unpublished rate is unknown, not zero. Workspace-specific pricing, taxes, balance-purchase fees, billing adjustments, and tool charges can affect the amount paid.

| Usage | Published base rate | Billing unit |
| --- | --- | --- |
| Input | $0.68 | per million input tokens |
| Output | $2.09 | per million output tokens |
| Cache read | $0.07 | per million cached input tokens |
| Cache write (default retention) | Not published | per million tokens |
| Cache write (one-hour retention) | Not published | per million tokens |
| Audio input | Not published | per million audio input tokens |

## Benchmarks

Last checked: 2026-10-14T17:57:48.915Z.
Configurations are evaluated separately. API scores retain their original units. Unmeasured scores are omitted.

### Mistral Large 4 Preview

Source: [Evaluation source](<https://artificialanalysis.ai/models/mistral-large-4>)

| Benchmark | Score |
| --- | --- |
| Intelligence Index | 38.38 |
| Omniscience Index | -5.30 |
| Humanity’s Last Exam | 35.03% |
| SciCode | 54.17% |
| CritPt | 10.57% |
| AA-LCR | 81.33% |
| Harvey LAB | 3.06% |
| GDPval Elo | 1422.17 |
| GDP.pdf All-pass | 18.60% |
| AutomationBench Partial | 59.90% |
| MMMU-Pro | 76.42% |

Full benchmark data and breakdowns: [Complete benchmark snapshot](<https://api.impossibl.com/v1/models/mistral/mistral-large-4/benchmarks>).


## Availability and retirement

The model is listed as serving. No retirement is announced in the current public catalog.

## Frequently asked questions

### What is Mistral Large 4?

Mistral Large 4 is Mistral's open-weight, general-purpose multimodal model, a granular mixture-of-experts with 52B active and 1.05T total parameters plus a 1.6B vision encoder. It takes text and image input over a long context, can reason before answering or answer directly, and supports function calling and structured outputs.

Sources: [Mistral Large 4 model card](<https://docs.mistral.ai/models/mistral-large-4-0>); [Mistral changelog](<https://docs.mistral.ai/resources/changelogs>); [Mistral reasoning documentation](<https://docs.mistral.ai/studio/conversations/reasoning>)

### How much does Mistral Large 4 cost?

Mistral Large 4 costs $0.68 per million input tokens and $2.09 per million output tokens through the Impossibl API. Cache reads cost $0.07 per million tokens. All rates are in USD.

### What is the context length of Mistral Large 4?

Mistral Large 4 has a 524,288-token context window through the Impossibl API.

### What inputs and outputs does Mistral Large 4 support?

Mistral Large 4 accepts text and image as input. Mistral Large 4 returns text.

### How do I use Mistral Large 4 through an API?

Mistral Large 4 is available through the Impossibl API. Use mistral/mistral-large-4 as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request.

### When was Mistral Large 4 released?

Mistral released Mistral Large 4 on October 6, 2026.

Sources: [Creator release announcement](<https://docs.mistral.ai/resources/changelogs>)

## Sources and related links

Prices, limits, capabilities, endpoints, and availability come from the public Impossibl catalog. Creator documentation supplies descriptions and release dates where available; it does not override the published API offering.

- [Public Impossibl model catalog](<https://api.impossibl.com/v1/models>)
- [Full model directory as JSON](<https://impossibl.com/models.json>)
- [Mistral Large 4: model details and pricing](<https://impossibl.com/mistral/mistral-large-4>)
- [Model directory](<https://impossibl.com/models>)
- [Mistral Large 4 model card](<https://docs.mistral.ai/models/mistral-large-4-0>)
- [Mistral changelog](<https://docs.mistral.ai/resources/changelogs>)
- [Mistral reasoning documentation](<https://docs.mistral.ai/studio/conversations/reasoning>)