lmstudio

LM Studio: per-model defaults

model_defaults

Idempotent

Save and manage per-model load defaults such as context length, TTL, GPU offload, and parallel slots, and view or clear them when not specified elsewhere.

Instructions

Save default load settings for a model: context length, TTL, GPU offload, parallel slots. load_model, chat, sessions and tasks use them whenever a setting is not given. With only model it shows that model's defaults, without arguments all of them; clear: true removes them.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "model_defaults"
  }
}

Input Schema

Name	Required	Description
`gpu`	No	GPU offload: "off", "max" or a ratio such as "0.5".
`ttl`	No	Unload automatically after this many idle seconds.
`clear`	No	Remove the saved defaults of this model.
`model`	No	Model key from list_models.
`parallel`	No	How many predictions may run at the same time.
`context_length`	No	Context window to load the model with, e.g. 262144.

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Nov 11, 2026

TDQS

A4.2/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds genuinely non-obvious behavior beyond the annotations: the saved defaults are applied implicitly as fallbacks by load_model, chat, sessions and tasks whenever a setting is omitted, and saving is partial (only the settings named). Annotations already cover safety (idempotent, non-destructive), so the remaining gap is what the read/merge output looks like. The 'clear: true' removal is mitigated by it being an explicit opt-in flag, so it does not contradict destructiveHint=false.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three tight clauses that together define the write form, the cross-tool effect, and the read and clear forms. Nothing is repeated and the primary purpose leads the sentence.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With six optional parameters, full schema coverage, annotations, and no output schema, the description covers everything needed to invoke the tool correctly. The one omission is that the shape of a returned defaults map is never hinted at, though that is minor given no output schema exists.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so each parameter is already documented in the schema, establishing a baseline of 3. The description adds mode meaning not present in the schema, namely that 'model' alone selects the read-one form and no arguments selects the read-all form, but per-parameter detail such as TTL or parallel semantics is left entirely to the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Save default load settings for a model') and enumerates the settings (context length, TTL, GPU offload, parallel slots). It also names the consumers ('load_model, chat, sessions and tasks use them'), which separates this configuration tool from the loading/chat tools in the sibling list.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives usable when-to-use guidance by mode: only 'model' → read one model's defaults, no arguments → list all, 'clear: true' → remove. The one gap is that it never states an explicit exclusion, such as a pointer to load_model for actually loading with these settings now.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.