Mistral Large 3 · Mistral

What does Mistral Large 3 cost to run in a coding agent?

Mistral Large 3 costs about $0.012 per average coding-agent request on our measured traffic, at list prices of $0.50 per million input and $1.50 per million output tokens. It is an open-weight generalist with EU provenance and scores 9 on the Artificial Analysis Intelligence Index. Source: Artificial Analysis (artificialanalysis.ai)

List prices: Mistral API pricing, checked 2026-10-13. Request costs: Standard Compute gateway data, 79,301 coding-agent requests from 75 users, Sep 13 to Oct 12, 2026. How we compute this

A newer version is out. Mistral Large 4 →

Run Mistral Large 4 in your agent

Or let the router send each step to the cheapest model that can do it, on providers we vet for price, uptime and speed. Plans from $19 a month.

See plans

What Mistral Large 3 costs per coding-agent request

Cost at list price	Mistral Large 3
Prompt under 100K tokens50% of requests, 87% of input cached	$0.0071
Prompt of 100K to 272K39% of requests, 88% of input cached	$0.018
Prompt over 272K tokens11% of requests, 86% of input cached	n/a
Average request	$0.012
100 requests	$1.19
Median session16 requests; a session ends after 10 idle minutes	$0.14
A month of regular usemedian of 20 users active on 5+ days	$15.10

Each logged request priced at the vendor’s list rates for its own token counts, including cache reads and any higher rate for long prompts. Average output: 698 tokens per request. 12% of requests were larger than Mistral Large 3's 262K context window and are left out of its figures.

Mistral Large 3 benchmarks, speed and latency

Artificial Analysis	Value	Rank
Intelligence Index10 evaluations, 0 to 100	9	47 of 48
Terminal-Bench 4.0agentic coding in a terminal	0.0%	35 of 39
Output speedmedian	73 tokens/s	29 of 44
Time to first answer tokenincludes thinking time	1.1 s	1 of 44

Source: Artificial Analysis (artificialanalysis.ai). Measured variant: Mistral Large 3, speed on Mistral. Fetched October 13, 2026. Rank is among the models on this site, older versions included. Artificial Analysis runs Terminal-Bench 4.0 in its own setup, so its scores differ from the tbench.ai leaderboard.

About Mistral Large 3

Mistral Large 3 (2512) is Mistral's open-weight general model, released December 2, 2025 with a 262K context. Its EU-based provenance matters for some compliance needs. It is a non-reasoning model that scores 9 on the Artificial Analysis Intelligence Index and 0% in Artificial Analysis's Terminal-Bench 4.0 run, far below current coding models, and Artificial Analysis lists it as superseded. Its 262K context is smaller than 11.7% of the coding-agent requests we measured.

Sources: Artificial Analysis

Mistral Large 3 FAQ

How much does Mistral Large 3 cost per coding-agent request?

$0.012 on average at list price, $1.19 per 100 requests, and $0.14 for a median 16-request session. That is 79,301 real coding-agent requests (Sep 13 to Oct 12, 2026) priced at Mistral's list rates; the median request sent 99,534 input tokens, 87% of input was read from cache, and requests returned 698 output tokens on average.

Does Mistral Large 3 charge more for long prompts?

No. Mistral Large 3 bills the same rate per token up to its 262K-token context window, which is smaller than 12% of the coding-agent requests we measured.

How fast is Mistral Large 3?

Artificial Analysis measures Mistral Large 3 at 73 output tokens per second, with 1.1 seconds to the first answer token, thinking time included, on Mistral's API. Source: Artificial Analysis (artificialanalysis.ai).

Is Mistral Large 3 good for AI agents?

General open-weight text work where EU provenance matters. On our scoring, which is calibrated to public benchmarks, it rates 4.4/10 for agentic ability and 3.4/10 for output quality. Open weights.

What is Mistral Large 3's context window?

Mistral Large 3 has a 262K-token context window. Mistral is the maker.

Is Mistral Large 3 open source?

Yes. Mistral Large 3 ships with open weights, so you can self-host it or run it through a hosted provider such as OpenRouter. Check the license before commercial use.

What are the downsides of Mistral Large 3?

The main trade-offs: Not suited to agentic coding (0% in Artificial Analysis's Terminal-Bench 4.0 run); 262K context, smaller than 11.7% of the coding-agent requests we measured; Superseded by Mistral Large 4 (October 2026). It's strongest for general open-weight text work where EU provenance matters.