LLMminimax

MiniMax: MiniMax M3 (batch)

MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...

Anyone in the Project can @-mention MiniMax: MiniMax M3 (batch) with the team's shared context - pooled credits, one chat, one memory.

All models

Starter is free forever - 1 Project, 100 credits/month, 1 MCP. No card.

Verdict

MiniMax M3 batch is a multimodal model with a massive 524K token context window and aggressive batch pricing at $0.15/$0.60 per Mtok — roughly 75% cheaper than comparable models. It handles text, images, and video in a single request, making it useful for document-heavy workflows where cost matters more than bleeding-edge performance. Without public benchmarks, you're trading proven accuracy for price and context capacity. Reach for this when you need to process large multimodal datasets on a tight budget and can tolerate higher latency from batch processing.

Best for

  • Batch processing of long documents with images
  • Cost-sensitive video content analysis
  • Large-scale multimodal data pipelines
  • Document extraction across mixed media types

Strengths

The 524K context window handles entire books, multi-hour video transcripts, or hundreds of pages with embedded images in one pass. Batch pricing undercuts most competitors by 3-4x, making it viable for high-volume jobs where per-token cost dominates your budget. Native video understanding eliminates the need to extract frames manually. The model accepts all three modalities in a single API call, simplifying workflows that mix PDFs, screenshots, and video clips.

Trade-offs

No public benchmarks means you're flying blind on accuracy relative to GPT-4o, Claude, or Gemini — expect to run your own evals before committing production traffic. Batch processing adds latency (typically minutes to hours), ruling out real-time use cases. MiniMax has less ecosystem maturity than OpenAI or Anthropic, so expect fewer integrations and community resources. The model is new enough that edge-case behavior and safety guardrails remain unproven at scale.

Specifications

Provider
minimax
Category
llm
Context length
524,288 tokens
Max output
471,859 tokens
Modalities
text, image, video
License
proprietary
Released
2026-05-31

Pricing

Input
$0.30/Mtok
Output
$1.20/Mtok
Model ID
minimax/minimax-m3:batch

Per-token prices show what the model costs upstream. On Switchy your team draws from one shared org credit pool - one plan, one balance for everyone.

Team cost calculator

Estimated monthly spend
$10.03
17.6M tokens / month
5 seats · 80 msgs/day

Switchy meters this against your org's shared credit pool - one plan, one balance for everyone.

Providers

Provider-level routing data is not available yet for this model.

Performance

Performance snapshots are collected daily. Check back after the next ingestion run.

Benchmarks

Public benchmark scores are not available yet for this model. Check back after the next ingestion run.

Works well with

Top MCPs

Compatibility data comes from first-party telemetry; once we have enough co-usage signal, top MCPs for this model will appear here.

How Switchy teams use it

Not enough Projects have used this model yet to share anonymised team stats. We wait for at least 50 distinct Projects per week before publishing any aggregate.

Starter prompts

Extract Tables from PDF

Extract all tables from the attached PDF as CSV. Preserve column headers exactly as shown. If a table spans multiple pages, merge it into one output. Include the page number where each table starts.
Open in a Project →

Summarize Video Content

Watch this video and create a summary with timestamps. For each major topic or scene change, note the time and write a 1-2 sentence description. Focus on key points and decisions.
Open in a Project →

Compare Document Versions

Compare these two document versions and list all substantive changes. Ignore formatting differences. For each change, quote the old text, the new text, and the page number. Flag any changes to images or diagrams.
Open in a Project →

Analyze Mixed-Media Reports

Read this report and extract: (1) executive summary, (2) all numerical data from charts and tables, (3) descriptions of any photos or diagrams. Output as JSON with those three keys.
Open in a Project →

Batch Classify Support Tickets

Classify each support ticket into one of these categories: billing, technical, feature request, bug report. If a screenshot is attached, describe what it shows in 10 words or less. Output as a CSV with columns: ticket_id, category, screenshot_summary.
Open in a Project →

Compare with

More language models

See all language models

Data last verified 7 hours ago.Sources aggregated hourly to weekly. See docs/architecture/model-directory.md.