This page is based on content/freetoken_en.md and reorganized into a cleaner guide for OpenClaw users who want to test free-token model workflows.
Why These Models Matter
Hunter Alpha and Healer Alpha were described in the draft as a high-value free pairing on OpenRouter, with both models exposed under the openrouter/ prefix and temporarily available at zero token cost for both input and output.
- Free-period pricing reported as
0 / 1Mfor both input and output - Hunter Alpha focuses on long-context planning and reasoning
- Healer Alpha focuses on multimodal execution with image and audio support
- The pair maps naturally to OpenClaw agent setups
Hunter Alpha
Hunter Alpha is the planning-heavy model in the pair.
Key Specs
- Claimed parameters:
1T - Context window:
1,048,576tokens - Max output:
32K - Speed: about
38 tokens/s - Latency: about
1.57s
Best Use Cases
- Large project planning
- Long-context reasoning
- Huge PDF or document analysis
- Complex codebase refactoring
- Agent-style multi-step workflows inside OpenClaw
Short Version
Hunter Alpha is the model to use when you want OpenClaw to think across a massive context window before acting.
Healer Alpha
Healer Alpha is the execution-heavy multimodal model in the pair.
Key Specs
- Context window:
262,144tokens - Max output:
32K - Speed: about
24 tokens/s - Latency: about
1.40s
Input Support
- Images via
image_url - Audio via
WAV - Video via
video_url - Text
Best Use Cases
- Screenshot and image analysis
- Audio transcription and summarization
- Cross-modal reasoning
- Real-world multimodal agent tasks
Short Version
Healer Alpha is the model to use when OpenClaw needs to see, hear, and act on more than plain text.
Best OpenClaw Split
- Use
Hunter Alphaas the planning brain - Use
Healer Alphaas the multimodal executor
That split gives OpenClaw a cleaner division between deep reasoning and real-world perception tasks.
How To Start
- Open
https://openrouter.ai - Sign up with email or Google
- Optionally create an API key in the dashboard
- Use chat directly at
https://openrouter.ai/chator connect the models inside OpenClaw
Model IDs
openrouter/hunter-alphaopenrouter/healer-alpha
Python Example
from openai import OpenAI
client = OpenAI(
base_url="https://openrouter.ai/api/v1",
api_key="your_free_key"
)
response = client.chat.completions.create(
model="openrouter/hunter-alpha",
messages=[{"role": "user", "content": "Your massive task..."}]
)
OpenClaw Note
The source draft says OpenClaw v13 added built-in installer options for both Hunter Alpha and Healer Alpha, so in compatible setups you may not need to manually type the model IDs at all.
Direct Links
- Hunter Alpha:
https://openrouter.ai/openrouter/hunter-alpha - Healer Alpha:
https://openrouter.ai/openrouter/healer-alpha - OpenRouter Chat:
https://openrouter.ai/chat - OpenClaw:
https://openclaw.ai