cape.agent.options.modelopts: Single-LLM options for CAPE agentic

The cape.agent.options provides tools to control options for the CAPE agent and the Large Language Models that it uses.

Here is a sample file:

{
    "ModelList": [
        "bartowski/Llama-3.2-3B-Instruct-GGUF",
        "meta-llama/Llama-3.1-8B-Instruct",
        "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16"
    ],
    "bartowski/Llama-3.2-3B-Instruct-GGUF": {
        "ToolSet": "small",
        "SkillSet": "none",
    }
}

The MaxToolCallLoops option (defaults based on ToolSet) controls how many rounds of tool calling the agent can perform. The Vision option (default true) controls whether tools that return images, such as cape.agent.tools.cntltools.view_subfig(), are exposed to the model:

{
    "ModelList": ["my-model"],
    "my-model": {
        "ToolSet": "medium",
        "MaxToolCallLoops": 8,  // Override default of 6 for medium
        "Vision": false  // Text-only model; no image tools
    }
}
See Also:
class cape.agent.options.modelopts.ModelOpts(*args, **kw)

Options structure, subclass of dict

Call:
>>> opts = ModelOpts(fname=None, **kw)
Inputs:
fname: str

File to be read as a JSON file with comments

kw: dict

Options added into opts

Outputs:
opts: Options

Options interface

get_opt(opt: str, **kw)

Get option, deriving MaxToolCallLoops from ToolSet if needed