Product Launch

MiniMax M3 Release Watch: What Developers Should Know Before API Access

Name: EvoLink AI Model API Platform
Brand: EvoLink
Availability: InStock

EvoLink Team

Product Team

May 31, 2026

7 min read

MiniMax M3 has started to attract developer attention after public discussion described it as a new-generation LLM for coding agents, long-context workflows, multimodal reasoning, and cost-efficient production use.

For teams building on EvoLink, the important question is practical: can you call MiniMax M3 through an API today, and should you plan production workloads around it yet?

As of May 31, 2026, the answer is: almost. MiniMax M3 is scheduled to go live on EvoLink on June 1, 2026. Developers will be able to access M3 through the same OpenAI-compatible endpoint used for all other EvoLink models.

MiniMax M3 status at a glance

Topic	Status as of June 1, 2026	What it means for developers
Public release signal	Confirmed	M3 is launching on EvoLink
EvoLink route availability	Going live June 1, 2026	Developers can access M3 through EvoLink
Official MiniMax model docs	M3 not listed in docs checked	Official MiniMax docs may follow separately
Model ID	To be confirmed on launch	Check EvoLink model catalog on June 1
Pricing	To be confirmed on launch	EvoLink pricing will be available at launch
Context length	Reported, pending final confirmation	Check model page after launch
Multimodal support	Reported, pending final confirmation	Wait for API schema and modality docs
Hugging Face / open model status	Not listed in the official MiniMaxAI models checked	Do not assume weights or license terms

This page will be updated when official MiniMax documentation or EvoLink pricing details are confirmed.

Why developers are watching MiniMax M3

The interest around MiniMax M3 is easy to understand. The public signal frames M3 around several things production AI teams care about:

Coding and agentic workloads where models need to plan, edit, call tools, and recover from mistakes.
Long-context tasks such as full codebase analysis, large contracts, long documents, and multi-file reasoning.
MiniMax Sparse Attention (MSA) as a reported architecture direction for handling very long context more efficiently.
Native multimodal reasoning for computer-use agents and product interfaces.
Lower-cost frontier-model routing if the reported pricing position is confirmed.

These are exactly the kinds of workloads where a unified API gateway matters. A team may want to evaluate M3 when it becomes available, but it also needs fallback options, cost controls, and a way to switch models without rewriting application code.

What has been reported but still needs confirmation

The strongest public signal so far is a social post attributed to Skyler Miao describing M3 as a new-generation LLM with MiniMax Sparse Attention for coding and agentic tasks. EvoLink treats that as a demand signal, not as final technical documentation.

Reported claim	What needs official confirmation
MiniMax Sparse Attention architecture	MiniMax technical docs or release notes
SOTA coding and agentic performance	Official benchmarks plus independent production-style evaluation
1M context	Official context, output, and billing limits
Native multimodal reasoning	API schema and supported input/output modalities
DeepSeek-like pricing	Official token pricing or plan pricing
Lower cost than Sonnet or other open models	Source-backed pricing comparison with exact model versions and dates
Open-model positioning	Official repository, model weights, and license terms

This distinction matters. Developers can prepare for M3 based on the signal, but production teams should not hardcode model IDs, quote pricing, publish benchmarks, or design around 1M context until official sources confirm those details.

API availability, model ID, and pricing

For API users, the missing pieces are straightforward:

Item	Current status	Why it matters
API availability	Not publicly listed in checked MiniMax model docs	Determines whether developers can call M3
Model ID	Not publicly listed	Required for SDK calls, routing config, and examples
Pricing	Not publicly listed in checked pricing pages	Required for budget planning and cost comparison
Rate limits and regions	Not publicly listed	Required for production rollout planning
Streaming and tool behavior	Not publicly listed	Required for coding agents, CLIs, and long-running workflows

MiniMax M3 is scheduled to launch on EvoLink on June 1, 2026. These details will be confirmed at launch — check the EvoLink model catalog for the latest information.

What EvoLink users can do now

MiniMax M3 is going live on EvoLink on June 1, 2026. Here is how to prepare:

Check the EvoLink model catalog on June 1 for the confirmed model ID, pricing, and context limits.
Keep your existing OpenAI-compatible integration — M3 will work through the same EvoLink endpoint and SDK setup.
Prepare a test set for coding-agent and long-context prompts so you can evaluate M3 quickly once it is live.
Keep fallback routes ready — for MiniMax-family workloads before M3 goes live, MiniMax-M2.5 on EvoLink remains a confirmed route for coding agents, repo Q&A, and long-context workflows.

When MiniMax M3 is ready to evaluate

MiniMax M3 launches on EvoLink on June 1, 2026. Once live, confirm these details from the EvoLink model catalog:

Official model ID
Input, output, and cache pricing
Context and max output limits
Supported modalities
Streaming, tool use, and function calling support
Rate limits and region availability

FAQ

Is MiniMax M3 released?
MiniMax M3 is scheduled to go live on EvoLink on June 1, 2026.

Is there a MiniMax M3 API?
Yes — MiniMax M3 will be accessible through the EvoLink API starting June 1, 2026, using the same OpenAI-compatible endpoint.

What is the MiniMax M3 model ID?
The model ID will be confirmed when M3 goes live on June 1. Check the EvoLink model catalog for the official ID.

How much does MiniMax M3 cost?
Pricing will be available on the EvoLink model catalog starting June 1, 2026.

Does MiniMax M3 support 1M context?
1M context has been reported in public discussion. Official context and output limits will be confirmed at launch.

Does MiniMax M3 support multimodal reasoning?
Native multimodal reasoning has been reported. Check the EvoLink model page after launch for confirmed modality support.

Does EvoLink support MiniMax M3?
Yes — MiniMax M3 is launching on EvoLink on June 1, 2026.

What should I use before MiniMax M3 goes live?
For MiniMax-family workflows, start with MiniMax-M2.5 on EvoLink. For coding-agent routing more broadly, compare confirmed models through EvoLink and keep fallback routes available.