Home › Compare › MiniMax: MiniMax M2-her vs MiniMax: MiniMax M3
MiniMax: MiniMax M2-her vs MiniMax: MiniMax M3: API Pricing Compared
Comparing MiniMax: MiniMax M2-her and MiniMax: MiniMax M3: on combined input+output list price, MiniMax: MiniMax M2-her is cheaper at $1.5 vs $1.5 per 1M tokens — a difference of $0 (free) (0%). MiniMax: MiniMax M2-her costs $0.3 in / $1.2 out with a 65,536-token context; MiniMax: MiniMax M3 costs $0.3 in / $1.2 out with a 1,048,576-token context.
| MiniMax: MiniMax M2-her | MiniMax: MiniMax M3 | |
|---|---|---|
| Input /1M | $0.3 | $0.3 |
| Output /1M | $1.2 | $1.2 |
| Cached input /1M | $0.03 | $0.06 |
| Context window | 65,536 | 1,048,576 |
| Max output tokens | 2,048 | 512,000 |
| Cost @ 10M in + 2M out / mo | $5.4 | $5.4 |
Estimate your own workload in the cost calculator, or open the full pages for MiniMax: MiniMax M2-her and MiniMax: MiniMax M3.
Frequently asked questions
Which is cheaper, MiniMax: MiniMax M2-her or MiniMax: MiniMax M3?
MiniMax: MiniMax M2-her is cheaper on combined list price: $1.5 vs $1.5 per 1M tokens (input + output). At 10M input + 2M output tokens per month, MiniMax: MiniMax M2-her costs $5.4 and MiniMax: MiniMax M3 costs $5.4.
Which has the larger context window, MiniMax: MiniMax M2-her or MiniMax: MiniMax M3?
MiniMax: MiniMax M2-her has a context window of 65,536 tokens; MiniMax: MiniMax M3 has 1,048,576 tokens. The larger one is MiniMax: MiniMax M3.
Price snapshot October 21, 2026, refreshed daily.