Skip to main content
MiniMaxJun 12, 2026

MiniMax M3 Preview

minimax-m3-previewPrivateOpen weights

MiniMax-M3 preview is a 1.4T-parameter frontier model from MiniMax for coding, agentic workflows, and complex reasoning, served at fp8 with a 512K context window.

$0.30/$1.20
per 1M in / out
512K
context window
128K
max output
How it comparesamong 143 text models
GPQA Diamondhigher than 67%89.4%
Blended pricecheaper than 75%$0.52
Context windowlarger than 41%512K
Max outputlarger than 78%128K

Further right is better. Hover a dot for its model.

Time to first token— 
Output speed— 
Uptime— 

Last 24 hours of Venice traffic.How it’s measured

Capabilities

What the model takes in, returns and supports through the API.

InTextImageVideo
OutText
Streamingstream: true
Function callingtoolsReasoningAlways onVisionUp to 10 images
Video inputVideo in messages
Web searchenable_web_search
Prompt cachingReads at $0.06
Optimized for codeTuned for software work

Not supportedstructured outputs, audio input, X search and log probabilities

Benchmarks

GPQA Diamond
89.4%45th of 139 · MiniMax M2.5 scores higher for less

Specs

Model ID
minimax-m3-preview
Endpoint
POST /api/v1/chat/completions
Context window
524,288 tokens
Max output
131,072 tokens
Reasoning
Always on
Precision
FP8
Privacy
Private
Added
Jun 12, 2026

Compare

ModelInputOutputGPQA DiamondContextMax outputFirst tokenSpeed
MiniMax M3 Preview$0.30$1.2089.4%512K128K——
DeepSeek V4.1 Flash$0.30$1.2088.4%1M256K——
DeepSeek v4.1 Flash (1017)$0.30$1.2090.9%1M256K——
GPT-5.6 Luna$0.20$1.2088.4%1M128K——
GPT-5.6 Luna Pro$0.20$1.2086.5%1M128K——

Pricing

Billed per token in USD. Reasoning tokens bill as output.

Input$0.30per 1M tokens
Cached input$0.0680% off input
Output$1.20per 1M tokens

Estimate a workload

$0.0012per request
$12.00for 10,000 requests

Blended price against similar models

MiniMax M3 Preview$0.525

Per 1M tokens at three input tokens for each output token.

Quickstart

OpenAI-compatible: point any OpenAI SDK at the Venice base URL with a Venice API key.

curl https://api.venice.ai/api/v1/chat/completions \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "minimax-m3-preview",
"messages": [{ "role": "user", "content": "Explain TEE attestation in two sentences." }]
}'
Base URL https://api.venice.ai/api/v1

Parameters

ParameterValuesNotes
modelminimax-m3-previewRequired
messagestext · image_url · video_urlContent parts the model reads
max_completion_tokensUp to 131,072Counts reasoning tokens
toolsFunction definitionsWith tool_choice
streamtrue | falseServer-sent events
venice_parameters.enable_web_searchauto | on | offWeb results and citations