MiniMax: MiniMax M3
Analysis Summary
MiniMax M3 is MiniMax's flagship model, sitting at an intelligence index of 54.7, which places it in the top tier of the current landscape. It supports text, image, and video input, making it one of the most modality-rich models available. Its instruction following score (ifbench 0.83) is among the strongest in this batch, and long-context reasoning (lcr 0.74) is excellent for a model at this price point. Tool use and function calling are both supported.
For businesses, M3 is well suited to content generation, long-document analysis, multimodal workflows, and agentic pipelines. Its coding index of 43.4 is strong but not at the very top of the field, so pure software engineering teams may prefer a coding-specialist model. The video input capability is a genuine differentiator for media, marketing, and content teams processing video assets.
At $0.30 input and $1.20 output per million tokens, M3 offers outstanding value for its capability tier. Teams looking for a high-intelligence, multimodal model at low cost should consider it a primary candidate, particularly for content-heavy and agentic workloads.
Assessed June 6, 2026
Editorial notes
MiniMax M3 reaches near-frontier intelligence with strong instruction following, video and image input, a 1M token context, and very competitive pricing at $0.30 per million input tokens.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How MiniMax: MiniMax M3 compares
MiniMax: MiniMax M3 ranks #8 of 374 AI models we track for overall intelligence, #25 of 311 for coding, #29 of 286 for agentic tasks. Its 1M-token context window is larger than 97% of the models we list. At $0.30 per million input tokens it is cheaper than 43% of comparable models.
About MiniMax: MiniMax M3
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,..
Capabilities
Performance Indices
Source: Artificial Analysis
This model was released recently. Independent benchmark evaluations are typically completed within days of release, so these figures are preliminary and are likely to be updated as testing is finalised.
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does MiniMax: MiniMax M3 stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
minimax/minimax-m3
|
| Provider | minimax |
| Release Date | May 31, 2026 |
| Context Length | 1,048,576 tokens |
| Max Completion | 512,000 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.30 | $0.000300 |
| Output | $1.20 | $0.001200 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about MiniMax: MiniMax M3
How much does MiniMax: MiniMax M3 cost?
MiniMax: MiniMax M3 costs $0.30 per million input tokens and $1.20 per million output tokens.
What is the context window of MiniMax: MiniMax M3?
MiniMax: MiniMax M3 has a context window of 1,048,576 tokens (1M).
Is MiniMax: MiniMax M3 good for coding?
On our coding benchmark index, MiniMax: MiniMax M3 ranks #25 of 311 models, placing it in the top quartile of the field for code generation and debugging.
What can MiniMax: MiniMax M3 do?
MiniMax: MiniMax M3 supports image/vision input, tool use, and function calling.
Who created MiniMax: MiniMax M3?
MiniMax: MiniMax M3 is developed by MiniMax and was released on May 31, 2026.
