Models
zen4-coder
Code generation model with 480B (35B active) MoE and 262K context.
zen4-coder
Code Generation
The primary Zen code model. 480B total parameters with 35B active per token. Specialized for code generation, review, debugging, and agentic programming.
Specifications
| Property | Value |
|---|---|
| Model ID | zen4-coder |
| Parameters | 480B (35B active) |
| Architecture | MoE |
| Context Window | 262K tokens |
| Tier | ultra |
| Input Price | $3.60 / 1M tokens |
| Output Price | $3.60 / 1M tokens |
Capabilities
- Full-repository code understanding with 262K context
- Code generation across all major languages
- Code review and debugging
- Agentic programming with tool calling
- Multi-file refactoring
API Usage
curl https://api.hanzo.ai/v1/chat/completions \
-H "Authorization: Bearer $HANZO_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "zen4-coder",
"messages": [{"role": "user", "content": "Write a Go HTTP server with graceful shutdown."}]
}'from hanzoai import Hanzo
client = Hanzo(api_key="hk-your-api-key")
response = client.chat.completions.create(
model="zen4-coder",
messages=[{"role": "user", "content": "Write a Go HTTP server with graceful shutdown."}],
)
print(response.choices[0].message.content)See Also
- zen4-coder-pro -- Full BF16 precision for maximum accuracy
- zen4-coder-flash -- Faster, lighter code model
- Pricing -- Full pricing table