Start with PawClaw Free
The free model connects without your own key. Token quota is tracked separately for every agent.
- No API key
- Per-agent quota
- Free-token usage in the dashboard
Text, images, video, and voice
OpenClaw
Hermes Agent
Choose a model by task and see the cost before using it
Start with PawClaw Free. Stronger text and media models are optional, while availability and pricing stay synchronized with the provider catalog.
Free
tokens for every agent
Text
routine and advanced work
Media
images, video, and voice
Live
current catalog and pricing
Paid text models
Choose from the PawClaw-approved catalog. The active model appears first, and usage is charged to a separate AI balance.
- Current pricing
- Ranked discovery
- Actual charge history
Images and video
Connect a media model for a specific job. The agent uses the selected model instead of silently choosing an expensive replacement.
- Images
- Video
- Model price and parameters
Voice messages
Separate models transcribe voice and generate speech. New agents receiving an AI gift may have transcription enabled automatically.
- Transcription
- Speech generation
- Usage from AI balance
In-depth guide
How this works in real life
A practical look at capabilities, setup, and operational boundaries.
Choosing a model for the workload
Classification, short summaries, and simple drafts usually prioritize speed and predictable cost. Coding, long documents, and deep research benefit from stronger reasoning and a larger context window.
Do not choose a model by ranking or size alone. Prepare representative examples, compare accuracy, response time, and cost, then select the best fit for the production workflow.
- Estimate reasoning complexity
- Consider context size
- Compare real examples
- Measure speed and cost
Balancing quality and usage cost
A workflow can use different approaches at different stages. A fast model can handle preliminary sorting, while a stronger model reviews complex cases. This reduces cost without compromising quality where it matters.
Watch the growth of instructions and history: unnecessary context increases usage and can reduce precision. Keep only useful memory, remove repeated data, and rerun evaluation examples after switching models.
- Separate simple and complex stages
- Avoid unnecessary context
- Retest after changing models
- Monitor actual usage
Explore more
Related resources
Agent service and paid AI usage are tracked separately
Clear OpenClaw and Hermes Agent pricing with monthly service, free tokens, AI balance, paid models, media, and optional resources.
Read moreMessage your agent in a familiar channel
Talk to OpenClaw and Hermes Agent in Telegram, WhatsApp, web chat, and mobile apps with guided setup, QR pairing, or a short code.
Read moreYour agent runs recurring tasks on schedule
Run recurring OpenClaw or Hermes Agent briefings, checks, and workflows with time zones, manual test runs, and execution history.
Read moreFAQ
Frequently asked questions
Do I need to configure a server myself?
No. PawClaw.ai creates an isolated environment, launches the agent, and provides management from one dashboard.
Can I start without my own API key?
Yes. PawClaw connects a free model and allocates free tokens separately to each agent, so no custom key is required to start.