Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d271Policy30d66, 14 pendingSources1,015

Mistral Medium 3.5 128B

MistralReleased May 22, 2026Current

Score basis

Mistral Medium 3.5 128B is behind the leaders in long context, instruction following, coding, factuality, and multimodal tasks. Too few results yet to rate reasoning, agentic tasks, safety, math, or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Limited5 capabilities
  1. Long Context−26.2%1 of 3
  2. Instruction Following−29.8%1 of 3
  3. Coding−31.7%5 of 10
  4. Factuality−34.0%2 of 4
  5. Multimodal−34.2%1 of 6
Not rated5 capabilities

Too few results yet to rate Reasoning, Agentic, Safety, Math or Multilingual.

Price

$1.50input$7.50outputper million tokens

From Artificial Analysis · 3 providers tracked · All prices

Costs more than 75% of 330 priced models · 3:1 input-to-output blend, log scale

Evidence

30results on28benchmarks

  • 3 independently verified
  • 19 aggregator
  • 8 vendor-reported

From 5 sources · latest Oct 8, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Mistral Medium 3.5 128B benchmark results

30 results on 28 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Long Context

LimitedFull long context ranking

26.2% behind the leader1 of 3 ranked benchmarks measured

Instruction Following

LimitedFull instruction following ranking

29.8% behind the leader1 of 3 ranked benchmarks measured

Show 1 more instruction following resultHide 1 instruction following result

Coding

LimitedFull coding ranking

31.7% behind the leader5 of 10 ranked benchmarks measured

Factuality

LimitedFull factuality ranking

34.0% behind the leader2 of 4 ranked benchmarks measured

Multimodal

LimitedFull multimodal ranking

34.2% behind the leader1 of 6 ranked benchmarks measured

Show 1 more multimodal resultHide 1 multimodal result

Reasoning

Not enough dataFull reasoning ranking

0 of 6 ranked benchmarks measured

Agentic

Not enough dataFull agentic ranking

0 of 7 ranked benchmarks measured

Show 1 more agentic resultHide 1 agentic result

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 5 more resultsHide 5 results

Mistral Medium 3.5 128B: common questions

Who makes Mistral Medium 3.5 128B?

Mistral Medium 3.5 128B is made by Mistral.

When was Mistral Medium 3.5 128B released?

Mistral Medium 3.5 128B was released on May 22, 2026, according to Mistral's own announcement.

What is Mistral Medium 3.5 128B good at?

Mistral Medium 3.5 128B is behind the leaders in long context, instruction following, coding, factuality, and multimodal tasks. Too few results yet to rate reasoning, agentic tasks, safety, math, or multilingual tasks.

How much does Mistral Medium 3.5 128B cost?

Mistral Medium 3.5 128B costs $1.50 per million input tokens and $7.50 per million output tokens, according to Artificial Analysis. We track its price at 3 providers. At a mix of three input tokens to one output token, it costs more than 75% of the 330 priced models we track.

How many benchmarks has Mistral Medium 3.5 128B been tested on?

We track 30 results for Mistral Medium 3.5 128B on 28 benchmarks from 5 sources, 3 of them independently verified. The latest was recorded on Oct 8, 2026.

Which API features does Mistral Medium 3.5 128B support?

OpenRouter lists tool calling, structured outputs, and reasoning for Mistral Medium 3.5 128B.

About this record

Where Mistral Medium 3.5 128B's numbers come from, and every name it appears under.

Tracked since
Apr 30, 2026
Newest source mention
Sep 3, 2026

Where the results come from

Verification: 30 scores · 3 independently verified · 19 aggregator-attributed · 8 vendor-reported. How these tiers are assigned

From 5 sources on 3 sites. Artificial Analysis supplies 20 of them; the 3 independently verified results come from 2 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai20
  • api.llm-stats.com8
  • datasets-server.huggingface.co2

Also known as

mistral-medium-3.5mistralai/mistral-medium-3.5-128bunsloth/mistral-medium-3.5-128b-ggufmistral-medium-3-5