Launch promo: 30 days free for legacy models
Simple, transparent pricing
Simple billing. Cancel anytime. Early pricing is subject to adjustment after V4 launch.
The first month is a no-risk trial for legacy models so teams can validate quality, latency, and routing behavior. After the trial, pick a monthly tier to lock in your quota and rate limits. Usage above the subscription quota is billed with clear pay-as-you-go rates so you can scale without renegotiating a plan.
DeepSeek V4 access is separate and remains on waitlist until the official launch. We will publish V4 credits and pricing once release details are confirmed, while keeping existing subscription tiers stable for legacy models.
Lite
For solo builders and prototypes.
$9.90/month
- 10M tokens per month
- 200 images included
- 2 requests per second
- 1 API key
- Email support
Most popular
Plus
Best for growing teams.
$29.90/month
- 40M tokens per month
- 800 images included
- 5 requests per second
- 3 team seats
- 3 API keys
- Priority support
Pro
For power users and production workloads.
$59.90/month
- 120M tokens per month
- 2,500 images included
- 8 requests per second
- 5 team seats
- 5 API keys
- Priority support
Usage limits (trial)
Launch promo applies to legacy models for 30 days.
Daily rate limit: 100 requests per day.
Token limit: 200,000 tokens per day.
Image generation: 200 images per day (VL2/Janus).
Speed: 1 request per second (trial).
Pay-as-you-go rates
Overages are billed based on model usage.
V3.1 / Math-7B: $1.00 per 1M tokens
R1 (Reasoning): $1.50 per 1M tokens
VL2 / Janus: $0.02 per image (1.5x for HD)
Daily cap: $20 (adjustable in dashboard)
V4 coming soon
DeepSeek V4 access
Early access credits will be announced at launch. Pricing will be finalized after cost review.