Mistral Large 4
Mistral Large 4 is a reasoning model from Mistral released on Oct 6, 2026. It reads text and images and writes text, with a 1.05M token context window and up to 262K tokens per reply.
Show moreShow less
It costs $0.68 per million input tokens and $2.09 per million output tokens. Cached input bills at $0.07, 90% less than fresh input.
It supports tool calling, structured JSON output and prompt caching.
8 providers serve it, including Vercel AI Gateway, NanoGPT and Requesty.
On the Artificial Analysis Intelligence Index it scores 38.4, which places it at number 18 among the 43 models I track with independent scores.
- Modalities
- In / out price
- $0.68 / $2.09 per 1M
- Context
- 1.05M
- Max output
- 262K
- Released
- Oct 6, 2026
- Knowledge cutoff
- n/a
- Tool calling
- Structured output
- Reasoning
- Prompt caching
- Open weights
- Free to use
Mistral Large 4 pricing
USD, as listed on Oct 15, 2026. Most bills are input heavy, so the cached input rate matters as much as the headline price.
| Input | $0.68 / 1M tokens | |
|---|---|---|
| Output | $2.09 / 1M tokens | |
| Cached input (read) | $0.07 / 1M tokens | 90% off input |
Mistral Large 4 providers
8 providers serve it, including Vercel AI Gateway, NanoGPT and Requesty. Each host sets its own price and limits, so the cheapest one is not always the one with the longest context.
| Provider | Input /M | Output /M | Cache read /M | Context | Max output |
|---|---|---|---|---|---|
| Mistral First party | $0.68 | $2.09 | $0.07 | 1.05M | 262K |
| Vercel AI Gateway | $0.68 | $2.09 | $0.07 | 524K | 262K |
| NanoGPT | $0.68 | $2.09 | $0.07 | 524K | 262K |
| Requesty | $0.68 | $2.09 | $0.07 | 524K | 524K |
| LowRouter | $0.7112 | $2.19 | $0.0732 | 524K | 262K |
| DevPass (LLM Gateway) | $1.36 | $4.18 | $0.14 | 1M | 262K |
| LLM Gateway | $1.36 | $4.18 | $0.14 | 1M | 262K |
| Ollama Cloud | $1.36 | $4.18 | $0.14 | 1.05M | 262K |
What Mistral Large 4 costs per month
Pick a workload close to yours or type your own volumes.
Independent benchmarks
Measured by third parties, not reported by Mistral. Each links to its source and shows the date I checked it.
A composite of 10 independent evaluations (agents 30%, coding 20%, scientific reasoning 20%, general 30%), run by Artificial Analysis on each model, version 4.3.2.
Artificial Analysis, checked Oct 10, 2026
Human preference: people compare two anonymous answers and pick the better one. Ratings from the Text Arena overall board, style control on.
Arena (formerly LMArena), checked Oct 10, 2026
Benchmarks Mistral reports (9)
- deepswe-v1.1 (DeepSWE v1.1)61.7%
- swe-atlas-qna (SWE-Atlas-QnA)59.4%
- terminal-bench-4.0 (Terminal-Bench 4)28.3%
- coding-agent-index (Coding Agent Index (combined))49.8%
- aa-automationbench (AA AutomationBench, 657 business workflows)59.9%
- aa-briefcase (AA-Briefcase, long-horizon knowledge work (version not stated))1393
- cybench (Cybench, 40 security competition exercises)93%
- dense-200 (Dense 200, visual grounding)42%
- lakera-b3 (Lakera B3 AI Security Benchmark, attacks resisted)93.3%
Labs pick their own settings and harnesses, so compare these only within one lab.
Supported request parameters
What you can set when you call Mistral Large 4 through an OpenAI compatible API.
- Tools
tools - Structured outputs
structured_outputs - Reasoning
reasoning - Temperature
temperature
Alternatives to Mistral Large 4
Models that take the same inputs and call tools, closest in price, cheaper ones first.
- Index 22.2
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 24 providers.
by googleJul 9, 20261.05M context$0.30/M input tokens$2.50/M output tokens24 providers
- 1.05M context
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 32 providers.
by googleMar 20, 2025$0.30/M input tokens$2.50/M output tokens32 providers
- Index 25.2
Reads text, images and video. Tool calling, structured output, reasoning. Served by 24 providers.
by qwenJun 2, 20261M context$0.40/M input tokens$1.60/M output tokens24 providers
- 256K context
Reads text, images and video. Tool calling, structured output, reasoning. Served by 6 providers.
by bytedance-seedJun 23, 2026$0.50/M input tokens$2.50/M output tokens6 providers
- 262K context
Reads text and images. Tool calling, reasoning.
by bytedanceJun 23, 2026$0.50/M input tokens$2.50/M output tokens
- 262K context
Reads text and images. Tool calling, structured output, reasoning.
by nex-agiJun 2, 2026$0.50/M input tokens$2.50/M output tokens
More from Mistral
- Index 14.2
Reads text, images and files. Tool calling, structured output, reasoning. Served by 18 providers.
by mistralaiApr 29, 2026262K context$1.50/M input tokens$7.50/M output tokens18 providers
- 262K context
Reads text and images. Tool calling, structured output, reasoning. Served by 5 providers.
by mistralaiApr 29, 2026$1.50/M input tokens$7.50/M output tokens5 providers
- Index 11.3
Reads text and images. Tool calling, structured output, reasoning. Served by 15 providers.
by mistralaiMar 16, 2026262K context$0.15/M input tokens$0.60/M output tokens15 providers
- 262K context
Reads text and images. Tool calling, structured output, reasoning. Served by 4 providers.
by mistralaiMar 16, 2026$0.15/M input tokens$0.60/M output tokens4 providers
Voxtral Mini TTS
SpeechTurns text into spoken audio.
by mistralaiMar 1, 2026
Reads audio.
by mistralaiFeb 1, 2026
Mistral Large 4 questions
How much does the Mistral Large 4 API cost?
$0.68 per million input tokens and $2.09 per million output tokens, with cached input at $0.07. The estimator on this page turns that into a monthly bill for your own workload.
What is the context window of Mistral Large 4?
1,048,576 tokens of input, and it can write up to 262,144 tokens in one reply. As a rough guide, a million tokens is around 750,000 English words.
Which providers serve Mistral Large 4?
8 providers serve it, including Vercel AI Gateway, NanoGPT and Requesty. Prices differ by host, so the Providers table on this page lists each one with its input and output price.
Does Mistral Large 4 support tool calling?
Yes. It accepts tool definitions and can return JSON that matches a schema, which is what an agent needs to call your APIs.
Is Mistral Large 4 open source?
No. Mistral serves it through an API only, so you cannot download the weights or host it yourself.
When was Mistral Large 4 released?
Oct 6, 2026.
How good is Mistral Large 4 compared with other models?
It scores 38.4 on the Artificial Analysis Intelligence Index and 1429 on the Arena text leaderboard. Benchmarks are a shortlist, not a verdict: test the top two or three candidates on your own tasks before you commit.
Catalog data from models.dev (MIT licensed), refreshed daily, last on Oct 15, 2026. Independent scores from my hand checked rankings, sources linked above. The summary is written from these facts.
Building on Mistral Large 4?
I test your real tasks on it and the two closest alternatives, then ship the agent on whichever wins on cost and accuracy.
Book Meeting