The Neural Net

search_models

Search the model register: open-weight AI model releases with exact parameter counts, architectural context length, licence and release date. Use this to answer questions about which models exist, how big they are, what they are licensed under, or what shipped recently.

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": "search_models"
  }
}

Input Schema

Name	Required	Description
`task`	No	e.g. text-generation, image-text-to-text, mask-generation
`limit`	No	Default 25, max 100
`query`	No	Substring match on the model id
`license`	No	e.g. apache-2.0, mit
`reasoning`	No	Only models whose chat template shows reasoning/thinking support
`image_input`	No	Only models that accept image input
`min_context`	No	Minimum architectural context length in tokens
`organisation`	No	e.g. qwen, deepseek-ai, meta-llama, google
`tool_calling`	No	Only models whose chat template shows tool-calling support
`max_parameters`	No
`min_parameters`	No	Minimum parameter count, e.g. 7000000000
`released_after`	No	ISO date, e.g. 2026-01-01
`include_derivatives`	No	Include quantisations and conversions of other releases (GGUF, MLX, AWQ and similar). Default false.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 20, 2026

changedInput schema / properties / task / description

Previous value: -"e.g. text-generation, image-text-to-text"New value: +"e.g. text-generation, image-text-to-text, mask-generation"

First observedOct 18, 2026

TDQS

A3.8/5.0

Behavior3/5

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

No annotations are provided, so the description carries the full burden. It does disclose the record shape (parameter counts, architectural context length, licence, release date) and thereby implies a read-only, list-returning operation, but it says nothing about result ordering, pagination behaviour beyond the schema's limit field, or whether filters combine with AND. Solid on content coverage, thin on behaviour.

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?

Two sentences, zero waste. The subject matter is front-loaded and the usage clause follows immediately, so an agent gets the what and the when in one pass.

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

Completeness3/5

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

For a 13-parameter tool with no output schema and no annotations, the description covers purpose well but leaves real gaps: nothing about return shape, ordering, or how the many boolean filters interact. The 92% schema coverage absorbs most parameter risk, keeping this at minimum-viable rather than poor.

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

Parameters4/5

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

Schema description coverage is 92%, so the schema already documents nearly every parameter, and the baseline is 3. The description adds a little by tying the covered data (size, context, licence, dates) to the generation of questions the tool answers, but it does not explain filter mechanics such as AND-combination or what 'task' values are valid.

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

Purpose4/5

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

States a specific verb ('Search') and resource ('the model register'), then enumerates the covered fields: parameter counts, context length, licence, release date. An agent can tell it apart from get_model (single-record lookup) and register_stats (aggregates), but the description never names those siblings and its 'what shipped recently' phrasing overlaps with recent_releases, so it falls short of a 5.

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?

'Use this to answer questions about which models exist, how big they are, what they are licensed under, or what shipped recently' gives concrete intake questions an agent can pattern-match against. There is no explicit when-not guidance, and no routing to recent_releases or get_model for the overlapping cases.

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

All The Neural Net tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?