Models
Catalog
| Model | Max context length | Max output length | Reasoning | Tools |
|---|---|---|---|---|
| zai-org/GLM-5.3 | 256,000 | 128,000 | low, high, max | Yes |
| moonshotai/Kimi-K3 | 512,000 | 128,000 | low, high, max | Yes |
Descriptions
Model
The identifier to pass as model when calling the API, linked to the model's Hugging Face page.
Max context length
The maximum supported aggregated token length per request, including reasoning token count.
Max output length
The maximum supported output token length per request, including reasoning token count.
Reasoning
The model's reasoning effort. Higher efforts usually leads to better generation quality at the cost of longer reasoning context.
Check Inference for how reasoning_effort works.
Tools
The model's ability to use tool calling in an inference session.
To make an inference request, refer to the Quickstart Coding agent integration is shown in Harness Integration. API Documentation are shown here.