DeepSeek V3 vs MiniMax M2-her
Side by side from live data: prices, limits, who serves each model and how reliably, and how they do on our chess arena.
In short
- DeepSeek V3 is 1.2× cheaper at a typical mix of 3 input tokens per output token.
- DeepSeek V3 accepts longer prompts: 164K vs 66K tokens.
- DeepSeek V3's best provider was more reliable over 24 h: 99.51% vs 96.1%.
- MiniMax M2-her is the newer model (released 2026-01-23).
| DeepSeek V3 | MiniMax M2-her | |
|---|---|---|
| Vendor | DeepSeek | MiniMax |
| Released | 2024-12-26 | 2026-01-23 |
| Input / 1M | $0.257 | $0.30 |
| Output / 1M | $1.03 | $1.20 |
| Cached input / 1M | — | $0.03 |
| Blended / 1M (3:1) | $0.45 | $0.525 |
| Context | 164K | 66K |
| Max output | 16K | 2K |
| Accepts | text | text |
| Reasoning | No | No |
| Providers on OpenRouter | 3 | 1 |
| Cheapest provider, output | $0.89 | $1.20 |
| Best provider uptime, 24 h | 99.51% | 96.1% |
| Chess arena rating | not playing | not playing |
| Vendor API status | Operational | — |
Cost calculator
DeepSeek V3
$30.87 per month$1.03 per day
MiniMax M2-her
$36.00 per month$1.20 per day
DeepSeek V3 saves $5.13 a month on this workload.
At list prices per 1M tokens; prompt caching and batch discounts are not included.
Compare DeepSeek V3 with
Compare MiniMax M2-her with