MiniMax M3
FlagshipFlagship model with native multimodal input (image & video), 1M context, ideal for large codebases.
Need a China coding plan (GLM / MiniMax / Kimi / Volcengine / Xiaomi)? Email me at [email protected]
MiniMax Β· Token Plan
M3 flagship with native multimodal input and 1M context. Four tiers from $10 to $80/mo, high-speed M2.7 models reaching 100+ TPS. Annual billing gives 2 months free.
M3 and M2.7 share the same monthly token quota.
Flagship model with native multimodal input (image & video), 1M context, ideal for large codebases.
Shares monthly quota with M3. Standard ~50 TPS, high-speed ~100+ TPS. Reference: 100-2,000 prompts per 5h depending on tier.
Monthly token usage model. Annual billing = 2 months free.
Starter
$100/year
Beginners exploring AI coding
Plus
$200/year
Regular daily developers
Max
$500/year
Heavy professional users
Max HS
Top pick$800/year
Teams and latency-sensitive workflows
M3 understands images and videos in addition to text.
High-speed tier for responsive interactive coding.
Annual billing at $100-$800 saves two months vs monthly.
Plans use a monthly token-based quota. Starter ~100 prompts/5h, Plus ~300, Max and Max HS ~1,000. Actual consumption depends on context length.
Both allow ~1,000 prompts/5h. Max HS adds M2.7-highspeed at 100+ TPS for latency-sensitive workflows.
Yes. M3 has native multimodal input for images and video, sharing the same monthly quota as text models.
Annual plans cost $100 (Starter) to $800 (Max HS), effectively 2 months free compared to monthly billing.