Skip to content

Model Library

MistralReleased April 29, 2026

Mistral Medium 3.5

Mistral · Released April 29, 2026

Intelligence score

30

At a glance

Intelligence
30
Speed
140.3tokens/sec
Latency
2.02stime to first token
Input price
$1.50per 1M tokens
Output price
$7.50per 1M tokens
Context window
260Ktokens
Parameters
128billions

Summary

Comparison summary

Mistral Medium 3.5 scores 30 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9). Mistral Medium 3.5 generates output at 140.3 tokens per second (based on Mistral's API), which is above average compared to other open weight models of similar size (median: 88.2 t/s). Mistral Medium 3.5 costs $1.50 per 1M input tokens (at the higher end, median: $0.40) and $7.50 per 1M output tokens (at the higher end, median: $0.84), based on Mistral's API.

Specifications

Technical specifications

Model nameMistral Medium 3.5
ProviderMistral
Release dateApril 29, 2026
Intelligence score30
Speed140.3 tokens/sec
Latency (TTFT)2.02s
Input price$1.50 / 1M tokens
Output price$7.50 / 1M tokens
Context window260,000 tokens
Parameters128B
ReasoningYes
Input modalitiestext, image
Output modalities—
Open weightsYes
LicenseOther
API availability—
SourceModel Intelligence legacy profile: mistral-medium-3-5

Pricing

Token prices

Price per 1M tokens

Mistral Medium 3.5, as listed in the catalog record

Input $1.50 and output $7.50 per 1M tokens. Missing prices are hatched and marked "Not published".
  1. Input$1.50
  2. Output$7.50

Model Intelligence legacy profile: mistral-medium-3-5

FAQ · 20

Frequently asked questions

Mistral Medium 3.5 was released on April 29, 2026.

Mistral Medium 3.5 was created by Mistral.

Mistral Medium 3.5 scores 30 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9).

Mistral Medium 3.5 generates output at 140.3 tokens per second (based on Mistral's API), which is above average compared to other open weight models of similar size (median: 88.2 t/s).

Mistral Medium 3.5 has a time to first token (TTFT) of 2.02s (based on Mistral's API), which is somewhat higher than average compared to other open weight models of similar size (median: 1.52s).

Mistral Medium 3.5 costs $1.50 per 1M input tokens (at the higher end, median: $0.40) and $7.50 per 1M output tokens (at the higher end, median: $0.84), based on Mistral's API.

Mistral Medium 3.5 costs $1.50 per 1M input tokens and $7.50 per 1M output tokens (based on Mistral's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $1.16 per 1M tokens. Pricing may vary by provider.

When evaluated on the Intelligence Index, Mistral Medium 3.5 generated 91M output tokens, which is somewhat higher than average compared to other open weight models of similar size (median: 59M).

Yes, Mistral Medium 3.5 is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.

Mistral Medium 3.5 supports text and image input.

Mistral Medium 3.5 supports text output.

Yes, Mistral Medium 3.5 supports image input and can analyze, describe, and answer questions about images.

Yes, Mistral Medium 3.5 is multimodal. It can process text and image input and generate text output.

Mistral Medium 3.5 has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Mistral Medium 3.5 is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Mistral Medium 3.5 has 128 billion parameters.

Mistral Medium 3.5 is released under the Other license. This license allows commercial use.

Mistral Medium 3.5 achieves a score of 30 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Mistral Medium 3.5 is available via API through 1 provider.

Mistral Medium 3.5 is available through 1 API provider.