MiniMax M2.7 vs MiniMax M3
API pricing, context window and capabilities for MiniMax M2.7 and MiniMax M3, side by side. Prices are the cheapest serving route per 1M tokens, verified against LiteLLM and OpenRouter.
Verdict
computed from the published numbersMiniMax M3 wins on coding and on intelligence.
Price
Tied
Coding
MiniMax M3
Intelligence
MiniMax M3
Context window
Tied
Key takeaways
who wins on what, from the published numbersMiniMax M2.7 wins
- Cheaper output ($0.55 vs $0.96 /1M)
- Higher max output (1M vs 512K)
MiniMax M3 wins
- Cheaper input ($0.25 vs $0.23 /1M)
- More serving providers (5 vs 8)
- Supports vision
Cheaper overall
MiniMax M2.7
$0.25 + $0.55 /1M
Larger context
Tied
More providers
MiniMax M3
8 providers
Pricing & specs
cheapest serving route · per 1M tokens| MiniMax M2.7 | MiniMax M3 | |
|---|---|---|
| Input $/1M | $0.25 | $0.23 best |
| Output $/1M | $0.55 best | $0.96 |
| Context window | 1M | 1M |
| Max output | 1M best | 512K |
| Serving providers | 5 | 8 best |
| Released | 2026-03-18 | 2026-06-01 |
Benchmarks
Artificial Analysis indices, side by side| MiniMax M2.7 | MiniMax M3 | |
|---|---|---|
| Intelligence | 38.9 | 45.4 best |
| Coding | 52.6 | 58.6 best |
| GPQA Diamond | 87% | 93% best |
| Output speed | 66 t/s | 148 t/s best |
| Latency (TTFT) | 1.53s | 0.99s best |
Cost calculator
choose a serving route for each sideInput tokens / month
Output tokens / month
MiniMax M2.7
via DeepInfra
$2.25
MiniMax M3
via Wandb
$2.11
Capabilities
feature support, side by side| Feature | MiniMax M2.7 | MiniMax M3 |
|---|---|---|
| Vision | — | ✓ |
| Tool use | ✓ | ✓ |
| Reasoning | ✓ | ✓ |
| Audio | — | — |
| Prompt caching | ✓ | ✓ |
| Structured output | ✓ | ✓ |
| Web search | — | — |
Related comparisons
Frequently asked questions
Which is cheaper, MiniMax M2.7 or MiniMax M3?
MiniMax M3 is cheaper for input ($0.25 vs $0.23 per 1M tokens). MiniMax M2.7 is cheaper for output ($0.55 vs $0.96 per 1M tokens).
What is the context window difference between MiniMax M2.7 and MiniMax M3?
Both MiniMax M2.7 and MiniMax M3 have about a 1M-token context window.
How do MiniMax M2.7 and MiniMax M3 compare on benchmarks?
On the tracked Artificial Analysis indices, MiniMax M3 leads on coding, and MiniMax M3 leads on intelligence.
Pricing verified against LiteLLM + OpenRouter + provider pages · refreshed daily
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI