Usage Limits

Usage limits control how much you can interact with the models that LibrePartner offers over a specified time period. Think of this as your "budget" that determines how many messages you can send to the models, or how long you can talk with the models, before needing to wait for limit reset.

Your usage is affected by several factors: the length and complexity of conversation, features and tools that you use, and which model you are using. The different subscription plans (Plus, Heavy, Unlimited) have different usage allowances.

Plus Plan$18/month
Perfect for individual power users

Usage Limits

Every 5 hours:

  • Send around 50-100 prompts to tier 1 models
  • Send around 100-300 prompts to tier 2 models
  • Send around 300-600 prompts to tier 3 models

You can have three 5-hour sessions per day (resets at midnight UTC time).

Heavy Plan$60/month
For demanding workloads

Usage Limits

Every 5 hours:

  • Send around 200-400 prompts to tier 1 models
  • Send around 400-1,200 prompts to tier 2 models
  • Send around 1,200-2,400 prompts to tier 3 models

You can have three 5-hour sessions per day (resets at midnight UTC time).

Unlimited Plan$120/month
Unlimited usage for professionals

Usage Limits

Unlimited number of prompts and no session limit.

Subject to general restrictions and fair use policies.

General Restrictions
Important policies for all subscription plans

What this means:

  • You are not allowed to share your API keys or subscription with anyone
  • You are not allowed to use the API key in applications where other people than yourself have access
  • You are not allowed to spam our API endpoints with requests

Service Level & Reliability

Individual plans do not include uptime guarantees. As such, we recommend against using these API keys in production environments or for mission-critical applications where downtime would have significant consequences.

We also have limits on the number of concurrent requests which can be made.

Models & Model Tiers
Available models organized by performance tier

We offer a variety of models that you can use. We separate the models into three categories based on their capabilities and resource requirements.

Tier 1Premium Models
  • DeepSeek-V3.1-Terminus
  • GLM-4.6
Tier 2Balanced Models
  • GPT-OSS-120B
  • GLM-4.6 Air
Tier 3Efficient Models
  • GPT-OSS-20B