Lite Plan
For individuals and lean teams.
$15
Lightweight daily coding.
Baseline modelsStandard throughput
ModelsDeepSeek, Qwen, GLM etc.
Best forLight use
SupportStandard support
Model plans and pricing
Choose the right model access plan for your coding workflow. Lite covers everyday development with core models, while Pro expands into deeper model access and heavier delivery workloads across mainstream AI coding tools.
Pro Plan
For heavier workflows and complex delivery.
$80
Advanced coding and agents.
Baseline + advancedHigher throughput
ModelsClaude Code, ChatGPT, Gemini etc.
Best forHigh-intensity use
SupportPriority support
Detailed comparison
Detailed comparison
Compare Lite and Pro across pricing, model scope, task fit, request rate, token throughput, support, and team fit.
| Item | Lite | Pro |
|---|---|---|
| Price | $15 | $80 |
| Model focus | Domestic model coverage | Global frontier model coverage |
| Model examples | DeepSeek, Qwen, GLM, MiniMax, Moonshot | Claude Code, ChatGPT, Gemini, Mistral, xAI |
| Best-fit tasks | Autocomplete, code explanation, lightweight refactors | Complex reasoning, multi-agent collaboration, large-repo migrations |
| Request rate ceiling (RPM) | 20 RPM | 60 RPM |
| Token throughput ceiling (TPM) | 120,000 TPM | 500,000 TPM |
| Capacity tier | Standard throughput, suited to daily development and lightweight agent tasks | Higher throughput, suited to parallel agents, long-context analysis, and high-intensity development |
| Support | Standard activation and integration support | Priority support and faster expansion review |
| Best-fit teams | Solo developers, small teams, and lightweight coding workflows | Heavy users, engineering teams, and complex delivery environments |
Price
Lite
$15
Pro
$80
Model focus
Lite
Domestic model coverage
Pro
Global frontier model coverage
Model examples
Lite
DeepSeek, Qwen, GLM, MiniMax, Moonshot
Pro
Claude Code, ChatGPT, Gemini, Mistral, xAI
Best-fit tasks
Lite
Autocomplete, code explanation, lightweight refactors
Pro
Complex reasoning, multi-agent collaboration, large-repo migrations
Request rate ceiling (RPM)
Lite
20 RPM
Pro
60 RPM
Token throughput ceiling (TPM)
Lite
120,000 TPM
Pro
500,000 TPM
Capacity tier
Lite
Standard throughput, suited to daily development and lightweight agent tasks
Pro
Higher throughput, suited to parallel agents, long-context analysis, and high-intensity development
Support
Lite
Standard activation and integration support
Pro
Priority support and faster expansion review
Best-fit teams
Lite
Solo developers, small teams, and lightweight coding workflows
Pro
Heavy users, engineering teams, and complex delivery environments
How to start
Activation flow
Activation currently runs through operations and focuses on plan choice, model scope, quota, and rate limits.
01
Choose a plan
Pick Lite for everyday development or Pro for higher-volume, more complex work.
02
Confirm with operations
Align on model scope, monthly quota, RPM / TPM, and support expectations.
03
Receive setup details
After activation, receive your endpoint, API key, and model guidance.
Confirm the right access plan for your workflow
Share your coding tools, model needs, expected monthly volume, and support requirements to lock in Lite or Pro.