# Mistral Medium 3.5 price, benchmarks and context | Alher Tech

> Mistral Medium 3.5 by Mistral with its Artificial Analysis score (5, 53rd of 58), what each task costs and takes, its price per million tokens and its context.

- Canonical page: https://alhertech.com/en/ai-comparison/mistral-medium-3-5/
- Site: Alher Tech (custom software, websites and apps, ERP and CRM, AI agents and AI consulting, technology consulting and software maintenance, https://alhertech.com/)
- Contact: https://alhertech.com/en/contact/

---
# Mistral Medium 3.5.

Among the lowest scores, but one of the fastest per task.

Mistral · Open weights · Released 28 April 2026 Data from 13 November 2026.

- Score 53rd of 58, 5, Artificial Analysis's Intelligence Index. The median is 27.
- Cost 27th of 51, $1.04, What each task of that index costs. The median is $1.03.
- Time 9th of 50, 4.6 min, How long each task of that index takes. The median is 14.6 min.
- Context 48th of 55, 256 thousand, Tokens it reads at once, as 2 others do.

## In short

It scores 5 on Artificial Analysis's Intelligence Index, 53rd of the 58 models it scores. The median is 27.

Each task of that index costs it $1.04 and takes it 4.6 minutes on average. The median of the 58 is $1.03 and 14.6 minutes.

Mistral sells it at $1.50 per million input tokens and $7.50 per million output tokens, and it reads up to 256 thousand tokens at once.

Its weights are published, so it can run on your own servers.

## Its data

- **Maker**: Mistral

- **Weights**: Open

- **Effort measured**: Its only one

- **Price per million tokens**: $1.50 input and $7.50 output

- **Context**: 256 thousand tokens

- **Released**: 28 Apr 2026

The score, the cost and the time per task and its tests from Artificial Analysis, read on 13 November 2026 at its highest effort. The price, the context and the release date from Mistral's page. [The table of the 58](https://alhertech.com/en/ai-comparison/).

## Its tests

### Coding

- **FrontierCode 2.0**, 1.8%. Changes to real open-source projects. 46th of 58.

- **Terminal-Bench 4.0**, 0.0%. Complex work in the terminal, on its own. 48th of 58.

[Best AI for coding →](https://alhertech.com/en/ai-comparison/best-coding-ai/)

### Reasoning

- **HLE Diamond**, 4.7%. Expert-level questions. 50th of 58.

- **CritPt v2, hard subset**, 0.5%. Physics research problems. 54th of 58.

[Best AI for reasoning →](https://alhertech.com/en/ai-comparison/best-reasoning-ai/)

## Keep reading.

### [Which AI Model to Use to Build Your Company's Software](https://alhertech.com/en/blog/opus-5-5-vs-gpt-6-astra-vs-gpt-6-1-software-development/)

AI · Nov 13, 2026

### [Fundaia Case Study, Our Platform to Deploy Apps from GitHub](https://alhertech.com/en/blog/fundaia-case-study/)

Case study · Nov 12, 2026

### [Alquilia Case Study, an AI CRM That Qualifies Tenants over WhatsApp](https://alhertech.com/en/blog/alquilia-case-study/)

Case study · Nov 12, 2026

### [Natnatura Case Study, a Shopify Supplement Store in Spain and Portugal](https://alhertech.com/en/blog/natnatura-case-study/)

Case study · Nov 11, 2026

### [Noura Eclat Case Study, a Shopify Professional Skincare Store in English, Arabic and Spanish](https://alhertech.com/en/blog/noura-eclat-case-study/)

Case study · Nov 11, 2026

### [Claude Code Fixes Failed Deployments, and a Person Decides](https://alhertech.com/en/blog/claude-code-fixes-failed-deployments/)

AI · Oct 26, 2026

### [Leaderbrock Case Study, an Internal App for a Sales Team's Schedule, Expenses and Routes](https://alhertech.com/en/blog/leaderbrock-case-study/)

Case study · Nov 10, 2026

### [ReservPark Case Study, a Parking App for Street Spots and Hourly Garages](https://alhertech.com/en/blog/reservpark-case-study/)

Case study · Nov 10, 2026

### [Almacén-Pro Case Study, a Tool for Stores to Check Incoming Goods by Scanning](https://alhertech.com/en/blog/almacen-pro-case-study/)

Case study · Nov 11, 2026

### [How to Prevent Double Bookings](https://alhertech.com/en/blog/how-to-prevent-double-bookings/)

Software · Oct 25, 2026

### [Boathous Case Study, a Boat Rental Marketplace With a Website, Apps and AI](https://alhertech.com/en/blog/boathous-case-study/)

Case study · Nov 10, 2026

## Questions.

### How much does Mistral Medium 3.5 cost?

Mistral sells it at $1.50 per million input tokens and $7.50 per million output tokens. On the tasks of Artificial Analysis's index it costs $1.04 each.

### How much text does it read at once?

Up to 256 thousand tokens, according to Mistral.

### Can it run on your own servers?

Yes. Its weights are published, so it can be downloaded and run on your own servers.

### Which one for your company's software?

It depends on the job. We cover it in “[Which AI Model to Use to Build Your Company's Software](https://alhertech.com/en/blog/opus-5-5-vs-gpt-6-astra-vs-gpt-6-1-software-development/)”.

## Let's talk.

Tell us what you need and one of the partners who write the code gets back to you.

[Schedule a call](https://alhertech.com/en/book-a-call/)
