Skip to main content
Glama

Text To Music

text_to_music

Generate music from text description and lyrics.

Creates music tracks from text prompts with customizable parameters including tempo, key, time signature, and optional style reference audio.

IMPORTANT: Check model specifications using get_available_models() before calling. Pay attention to model limits for duration, inference_steps, guidance_scale, and bpm.


Available models:

  • MiniMaxMusic3_11B_INT8: steps=30 (fixed), guidance=1.7 (fixed), duration=20-300s, bpm=30-30, caption=3-2000 chars, ref_audio=0s (fixed)

  • AceStep_1_5_XL_Turbo_INT8: steps=8 (fixed), guidance=1 (fixed), duration=10-300s, bpm=50-200, caption=3-300 chars, ref_audio=5-60s

  • AceStep_1_5_Turbo: steps=8 (fixed), guidance=1 (fixed), duration=10-300s, bpm=50-200, caption=3-300 chars, ref_audio=5-60s

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "caption": "<caption>",
      "model": "<model>",
      "lyrics": "<lyrics>",
      "duration": "<duration>"
    },
    "name": "text_to_music"
  }
}

Input Schema

NameRequiredDescriptionDefault
bpmNoBeats per minute (optional). Bounds per model ('bpm' in the model list)
seedNoRandom seed (-1 for random)
modelYesMusic generation model name (must support txt2music inference type)
lyricsYesSong lyrics (max 10000 chars). Use '[Instrumental]' for no vocals
captionYesText description of the music to generate. Length bounds per model ('caption' in the model list)
durationYesAudio duration in seconds. Bounds per model ('duration' in the model list)
keyscaleNoMusical key/scale, e.g. 'C major', 'F# minor' (optional)
webhook_urlNoHTTPS URL for deAPI to call when this job finishes. Notifies your own systems; the tool still returns the result here
audio_formatNoOutput format, e.g. 'mp3' or 'flac'. Per-model — see info.limits.output_formats; leave unset for the model's own default
timesignatureNoTime signature: 2, 3, 4, or 6 (optional)
guidance_scaleNoGuidance scale. Leave unset when the model fixes it (see 'guidance' in the model list)
vocal_languageNoVocal language code, e.g. 'en', 'es' (optional)
webhook_secretNo32-255 char key deAPI signs the callback with (HMAC-SHA256). Needs webhook_url. A secret given in chat stays in the transcript
inference_stepsNoNumber of inference steps. Leave unset when the model fixes it (see 'steps' in the model list)
reference_audioNoReference audio for style transfer (URL, data URI, or base64). Its length must be within the model's window ('ref_audio' in the model list, e.g. 5-60s)

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.3/5.0
Behavior4/5

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

With no annotations, the description carries the full burden, and it delivers a concrete per-model constraint table (fixed steps/guidance, duration, bpm, caption and reference-audio windows) that an agent could not infer elsewhere. It is silent on job lifecycle behavior — whether the call blocks, how long generation takes, or when check_job_status applies — which is a meaningful gap for what is clearly an async generation job.

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

Conciseness4/5

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

Purpose and the critical pre-call warning are front-loaded, and the model table is dense rather than padded. There is mild redundancy in telling the agent to call get_available_models() and then inlining model specs that may go stale, plus the bpm range '30-30' for the first model looks like a data defect.

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?

For a 15-parameter generation tool with an output schema and no annotations, the description covers the hardest part — model-dependent valid ranges — well. It stops short of describing async execution, expected latency, or cost, and there is a separate text_to_music_price sibling the agent is never pointed to.

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 coverage is 100%, so the baseline would be 3, but the description genuinely adds meaning by enumerating each model's actual bounds for duration, bpm, caption length, inference_steps, and guidance_scale — the schema only says 'bounds per model'. It does not explain the webhook_url/webhook_secret pairing or reference-audio format options, which remain schema-only.

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?

The first sentence gives a specific verb and resource: 'Generate music from text description and lyrics', which cleanly separates it from text_to_audio, text_to_video, and other generation siblings. The follow-up sentence clarifies that parameters like tempo, key, time signature, and reference audio are customizable, confirming scope without ambiguity.

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?

It gives an explicit pre-call step — 'Check model specifications using get_available_models() before calling' — plus the specific parameter limits to watch. What's absent is any routing guidance between this tool and text_to_audio or the _price sibling, so an agent deciding which audio-generation tool to pick gets no help.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.