API route
high
Standard <=512K: $0.30/M input, $1.20/M output, $0.06/M cache read. Higher rates apply above 512K and priority tier.
Buyer-value caveat
PlanetGPU separates vendor-published prices from buyer guidance. API pricing, subscriptions, context windows, open-weight status, and referral terms can change quickly, so verify the linked official sources before purchase or infrastructure planning.
Context
1M target; 512K minimum noted by MiniMax
vendor source
Status
Hybrid
Verify weights
API pricing
high
transparency
Verified
2026-06-18
source freshness
Subscription route
Token Plan Plus $20/month, Max $50/month, Ultra $120/month.
Claude Code, Cursor, OpenClaw, TRAE, Hermes Agent, MiniMax CLI, MiniMax Code, Anthropic-compatible API, and OpenAI-compatible API.
GPU route
Verify weights
Treat self-hosting as a verification step until official weights, license terms, and serving requirements are confirmed.
Referral / affiliate offer
PlanetGPU may receive MiniMax referral vouchers if you subscribe through this link; MiniMax states invitees get 10% off and referrers receive vouchers equal to 10% of invitee payment.
Best for
Developers who want a subscription for M3 in Claude Code or Cursor-like workflows.
Teams needing a transparent API fallback for cost modeling.
Builders who want text, image, speech, music, and coding resources under one plan.
Caveats
- Token Plan quota is a shared usage bar with rolling and weekly windows, not a fixed public token bundle.
- Context above 512K input tokens has limited-availability notes in MiniMax pricing docs.
- Referral terms are promotional and should be verified at checkout.
Verification sources
Source freshness is stored per model opportunity and shown in the table and schema.