mistralai/mistral-large-4-0

模型发布 OpenRouter 模型 原文语言:中文

OpenRouter 于 2026-10-06 收录该模型(收录时间不等于模型正式发布日期);上下文:1,048,576 tokens;输入价:$0.7000/百万 tokens;输出价:$2.1000/百万 tokens;定位(来源描述):Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for…

全文

Mistral: Mistral Large 4

mistralai/mistral-large-4-0

Model weights

Compare

Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for reasoning, coding, and agentic workloads. It offers a 1M-token context window with up to 256K output tokens, and supports tool calling and structured outputs.

Modalities

In / out price

$0.70 / $2.10per 1M

Context

1.0M

Released

Oct 6, 2026

Compare

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput

$0.70$2.10$0.0701.54s70 tok/s

99.82%

$0.77$2.31$0.0771.29s80 tok/s

99.77%

ProviderInput / Output /M

Input $0.70 / Output $2.10

70 tok/s99.82% uptime

Input $0.77 / Output $2.31

80 tok/s99.77% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.172/M tokens

Output

$2.18/M tokens

Mistral Large 4 — Price history

Price history

No effective price data for this selection

$0.154$2.1086.7%88.1%

$0.357$2.3159.6%11.9%

ProviderInput / Output /M

Token share

Effective in $0.154 / Effective out $2.10

Token share

88.1%

Effective in $0.357 / Effective out $2.31

Token share

11.9%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

80tok/s

P50, best across providers

Latency

1.29s

P50, best provider

Reasoning effortLatency / throughput

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)The model was reachable. Request routed to a provider.

100.00%

Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.

87.81%

Availability over the last 3 days

Last 72 hours

Availability 87.81%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.84%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Mistral: Mistral Large 4 (Artificial Analysis, Design Arena and Vals AI)

SourceBenchmarkScore

Artificial AnalysisMistral Large 4 (High) Intelligence Index26.1

Artificial AnalysisMistral Large 4 (High) CritPt10.0%

Artificial AnalysisMistral Large 4 (High) Terminal-Bench 4.024.7%

Artificial AnalysisMistral Large 4 (High) AA-Omniscience Accuracy26.0%

Artificial AnalysisMistral Large 4 (High) AA-Omniscience Non-Hallucination Rate71.8%

Design ArenaMistral Large 4 Preview Models Arena 3D Elo1239

Design ArenaMistral Large 4 Preview Models Arena Asciiart Elo1197

Design ArenaMistral Large 4 Preview Models Arena Code Categories Elo1255

Design ArenaMistral Large 4 Preview Models Arena Data Visualization Elo1273

Design ArenaMistral Large 4 Preview Models Arena Game Development Elo1204

Design ArenaMistral Large 4 Preview Models Arena SVG Elo1222

Design ArenaMistral Large 4 Preview Models Arena UI Component Elo1278

Design ArenaMistral Large 4 Preview Models Arena Website Elo1258

Vals AIMistral Large 4 Preview BioMysteryBench67.0%

Vals AIMistral Large 4 Preview Code Migration30.6%

Vals AIMistral Large 4 Preview EMB v1.169.0%

Vals AIMistral Large 4 Preview Finance Agent v2.161.4%

Vals AIMistral Large 4 Preview Vals Finance Index56.9%

Vals AIMistral Large 4 Preview HLAB-Vals-1.117.5%

Vals AIMistral Large 4 Preview Immigration63.7%

Vals AIMistral Large 4 Preview IOI45.3%

Vals AIMistral Large 4 Preview Vals Legal Index65.6%

Vals AIMistral Large 4 Preview Legal Research Bench31.7%

Vals AIMistral Large 4 Preview MedCode40.7%

Vals AIMistral Large 4 Preview MedScribe80.4%

Vals AIMistral Large 4 Preview MysteryMechanism12.2%

Vals AIMistral Large 4 Preview ProofBench v1.110.0%

Vals AIMistral Large 4 Preview SWE-bench Multimodal33.1%

Vals AIMistral Large 4 Preview Tax Agent Bench30.4%

Vals AIMistral Large 4 Preview Terminal-Bench 4.027.8%

Vals AIMistral Large 4 Preview Vals Index48.7%

Vals AIMistral Large 4 Preview Vibe Code Bench 1-10014.3%

Vals AIMistral Large 4 Preview Vibe Code Bench v1.178.4%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

1.

Portkey AI

Control panel for AI apps

52.9Btokens

2.

Claude Code

Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.

38.5Btokens

3.

pi

There are many coding agents, but this one is yours.

19.5Btokens

4.

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

14.4Btokens

5.

omp

new

9.78Btokens

Rest of OpenRouter

7.99B

Top 5 Apps

7.84B

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Available through EU in-region routing

Mistral: Mistral Large 4 is available through EU in-region routing. Requests sent to eu.openrouter.ai are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).

Providers in EuropeMistralPrices in Europe

Mistral: $0.77/M tokens input, $2.31/M tokens output

Base URLhttps://eu.openrouter.ai/api/v1PlanRequires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing

Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs

Documentation

Guides, cookbooks, blog posts, and tools for building with Mistral: Mistral Large 4 on OpenRouter.

How to Benchmark LLMsBlog post

How to Stop an AI Agent Stuck in an Infinite LoopBlog post

Frequently asked questions

Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for reasoning, coding, and agentic workloads. It offers a 1M-token context window with up to 256K output tokens, and supports tool calling and structured outputs.

The Mistral Large 4 documentation on OpenRouter includes the How to Benchmark LLMs blog post and the How to Stop an AI Agent Stuck in an Infinite Loop blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.

Mistral Large 4 costs $0.70/M input tokens and $2.10/M output tokens, with separate rates for Cache Read at $0.07/M tokens.

Mistral Large 4 has a 1,048,576 token context window. It supports up to 262,144 completion tokens.

Yes. Mistral Large 4 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

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

Mistral Large 4 is served by 1 provider on OpenRouter: Mistral.

Mistral Medium 3.5, Mistral Small 4, Devstral 2 2512 and 16 more are other text models from Mistral.

Mistral Large 4 was released on October 6, 2026.

正文以来源语言显示。

参考来源

← 返回每日简报