MiniMax
Shanghai lab shipping the M-series text models alongside its Hailuo video and speech products.
Full MiniMax record →Head to head
MiniMax vs DeepSeek
MiniMax (MiniMax) and DeepSeek (DeepSeek) both sit in Writing. DeepSeek is the cheaper entry point at Pay-as-you-go API pricing based on token consumption. No flat monthly fee.. DeepSeek shipped more tracked changes in the last 30 days (9 vs 5). Every value below comes from the latest crawl of the vendors' own pages.
| Attribute | MiniMax MiniMax | DeepSeek DeepSeek |
|---|---|---|
| Segment | Writing | Writing |
| Version | M3 | DeepSeek-V4.1-Flash |
| Entry cost | from $0.3/org/mo | Pay-as-you-go API pricing based on token consumption. No flat monthly fee. |
| Pricing tiers | Free free · API input /M $0.3 · API output /M $1.2 | deepseek-flash (Input Cache Hit - Off-Peak) $0.003 · deepseek-flash (Input Cache Hit - Peak) $0.006 · deepseek-flash (Input Cache Miss - Off-Peak) $0.15 · deepseek-flash (Input Cache Miss - Peak) $0.3 · deepseek-flash (Output - Off-Peak) $0.6 · deepseek-flash (Output - Peak) $1.2 |
| Model stack | H3 · M2.5 · M2.7 · M3 · MiniMax Code · MiniMax H3 · MiniMax M2.5 · MiniMax M2.7 · MiniMax M3 · MiniMax Music 3.0 · MiniMax Speech 2.8 · MiniMax-M2.1 · Music 3.0 · Speech 2.8 | DeepSeek-R1 · DeepSeek-V3.2 · deepseek-v4-flash · DeepSeek-V4-Flash-0731 · DeepSeek-V4-Flash-Vision-Exp · DeepSeek-V4-Pro · DeepSeek-V4-Pro-0813 · DeepSeek-V4.1-Flash |
| Context window | 1000000 | 1M |
| Public API | ||
| Routes models | ||
| Changes / 30d | 5 | 9 |
| Origin | China | China |
| Capabilities | Interleaved thinking, Tool calling, Open weights, Long context, Text Generation, Code Generation, Video Generation, Speech Generation | Reasoning mode, Open weights (MIT), OpenAI-compatible API, Context caching, Function calling, Thinking Mode, Context Caching, JSON Output |
DeepSeek
Hangzhou lab whose V and R series models are released under permissive licences and undercut Western API pricing by an order of magnitude.
Full DeepSeek record →