Skip to content

Model Library

MistralReleased December 2, 2025

Ministral 3 14B

Mistral · Released December 2, 2025

Intelligence score

11

At a glance

Intelligence
11
Speed
78.2tokens/sec
Latency
860mstime to first token
Input price
$0.20per 1M tokens
Output price
$0.20per 1M tokens
Context window
260Ktokens
Parameters
14billions

Summary

Comparison summary

Ministral 3 14B scores 11 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 6). Ministral 3 14B generates output at 78.2 tokens per second (based on Mistral's API), which is below average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s). Ministral 3 14B costs $0.20 per 1M input tokens (somewhat higher than average, median: $0.15) and $0.20 per 1M output tokens (very competitive, median: $0.32), based on Mistral's API.

Specifications

Technical specifications

Model nameMinistral 3 14B
ProviderMistral
Release dateDecember 2, 2025
Intelligence score11
Speed78.2 tokens/sec
Latency (TTFT)0.86s
Input price$0.20 / 1M tokens
Output price$0.20 / 1M tokens
Context window260,000 tokens
Parameters14B
ReasoningNo
Input modalitiestext, image
Output modalities—
Open weightsYes
LicenseApache 2
API availability—
SourceModel Intelligence legacy profile: ministral-3-14b

Pricing

Token prices

Price per 1M tokens

Ministral 3 14B, as listed in the catalog record

Input $0.20 and output $0.20 per 1M tokens. Missing prices are hatched and marked "Not published".
  1. Input$0.20
  2. Output$0.20

Model Intelligence legacy profile: ministral-3-14b

FAQ · 20

Frequently asked questions

Ministral 3 14B was released on December 2, 2025.

Ministral 3 14B was created by Mistral.

Ministral 3 14B scores 11 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 6).

Ministral 3 14B generates output at 78.2 tokens per second (based on Mistral's API), which is below average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s).

Ministral 3 14B has a time to first token (TTFT) of 0.86s (based on Mistral's API), which is better than average compared to other open weight non-reasoning models of similar size (median: 1.54s).

Ministral 3 14B costs $0.20 per 1M input tokens (somewhat higher than average, median: $0.15) and $0.20 per 1M output tokens (very competitive, median: $0.32), based on Mistral's API.

Ministral 3 14B costs $0.20 per 1M input tokens and $0.20 per 1M output tokens (based on Mistral's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.20 per 1M tokens. Pricing may vary by provider.

When evaluated on the Intelligence Index, Ministral 3 14B generated 15M output tokens, which is somewhat higher than average compared to other open weight non-reasoning models of similar size (median: 9.5M).

No, Ministral 3 14B is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

Ministral 3 14B supports text and image input.

Ministral 3 14B supports text output.

Yes, Ministral 3 14B supports image input and can analyze, describe, and answer questions about images.

Yes, Ministral 3 14B is multimodal. It can process text and image input and generate text output.

Ministral 3 14B has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Ministral 3 14B is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Ministral 3 14B has 14 billion parameters.

Ministral 3 14B is released under the Apache 2.0 license. This license allows commercial use.

Ministral 3 14B achieves a score of 11 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Ministral 3 14B is available via API through 2 providers.

Ministral 3 14B is available through 2 API providers.