MiniMax M2.1 Lightning
import { streamText } from 'ai'
const result = streamText({ model: 'minimax/minimax-m2.1-lightning', prompt: 'Why is the sky blue?'})Copy link to headingPlayground
Try out MiniMax M2.1 Lightning by MiniMax. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
MiniMax M2.1 Lightning
MiniMax M2.1 Lightning
Copy link to headingProviders
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
Copy link to headingThroughput24 hours
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
Copy link to headingLatency24 hours
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Copy link to headingUptime24 hours
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
Copy link to headingMore models by MiniMax
| Model |
|---|
| 1M | 0.9s | 205tps | $0.60/M$0.30/M +1 more | $2.40/M$1.20/M +1 more | Read: $0.12/M$0.06/M+1 more Write: — | — | +1 |   | 05/31/2026 | ||||
| 205K | 0.5s | 163tps | $0.30/MFast $0.60/M | $1.20/MFast $2.40/M | Read:$0.06/M Write:$0.38/M | — |  | 03/18/2026 | |||||
| 205K | 1.1s | 48tps | $0.60/M | $2.40/M | Read:$0.06/M Write:$0.38/M | — | | 03/18/2026 | |||||
| 1M | 0.5s | 91tps | $0.07/MFast $0.60/M | $0.57/MFast $2.40/M | Read:$0.03/M Write:$0.38/M | — |    +3 | 02/12/2026 | |||||
| 205K | 0.4s | 146tps | $0.30/M | $1.20/M | Read:$0.03/M Write:$0.38/M | — | | 12/23/2025 | |||||
| 205K | 0.9s | 62tps | $0.30/M | $1.20/M | Read:$0.03/M Write:$0.38/M | — | | 10/27/2025 |
Copy link to headingAbout MiniMax M2.1 Lightning
MiniMax M2.1 Lightning shipped alongside M2.1 as its speed-optimized companion. The Lightning variant delivers faster inference while maintaining identical outputs to standard M2.1. You don't trade quality for throughput.
The model supports the same programming languages as M2.1: Go, C++, JavaScript, C#, TypeScript, Rust, Java, Kotlin, and Objective-C. It also carries the same Interleaved Thinking capability and agentic tool-use support that define the 2.1 generation.
Automatic prompt caching is built in with no manual configuration. This further reduces effective latency for repeated or structurally similar prompts. MiniMax M2.1 Lightning suits developer tools, IDE assistants, and any application where users expect near-instantaneous code suggestions.
Copy link to headingWhat To Consider When Choosing a Provider
- Configuration: For streaming use cases where time-to-first-token matters most, MiniMax M2.1 Lightning's throughput advantage translates directly into a more responsive end-user experience.
- Zero Data Retention: AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
Copy link to headingWhen to Use MiniMax M2.1 Lightning
Best for
- Interactive developer tools: IDE plugins where response latency is user-visible
- Real-time code completion: Inline suggestion features in web or IDE applications where latency is visible
- High-throughput batch jobs: Faster tokens-per-second reduces job duration
- Streaming user experiences: Applications that need low time-to-first-token
- Drop-in speed upgrade: Teams already on M2.1 who want faster inference
Consider alternatives when
- Minimize cost: Throughput is not a constraint, so use standard M2.1
- Architectural planning needed: Your tasks require the planning capabilities introduced in M2.5
- Vision input required: M2.1 Lightning is text-only, so use a multimodal model when your workload includes image inputs
Copy link to headingConclusion
MiniMax M2.1 Lightning resolves the typical quality-vs-speed tradeoff by matching M2.1's output while running faster. It's a straightforward upgrade for any latency-sensitive application already using the 2.1 generation. Built-in prompt caching amplifies the speed benefit for repetitive context patterns.