MiniMax Music Pro

Overview

MiniMax Music Pro (Music-2.6) converts text descriptions into complete audio files. It delivers higher fidelity and more complex arrangements than the base Music-2.0 variant, operating in the frontier-open quality tier. On Kyma, the model accepts text input and returns audio output with automatic request failover. Responses include exact generation costs in the usage object, and the X-Kyma-Model header confirms which variant executed the request. Repeated prompt prefixes are cached and billed at this model’s cached input rate. Generation latency is higher than turbo variants due to the focus on audio quality. The endpoint does not support structured outputs, vision, or reasoning, and it is strictly limited to music generation rather than speech synthesis or voice cloning.

Specs

Field	Value
Model ID	`minimax-music-pro`
Best for	Premium background music, brand soundtracks, podcast intros, video scoring
Context window	2,000
Input modalities	Text
Output modalities	Audio
Tool calling	Yes
Structured outputs	Yes
Prompt caching	Yes
Speed	Slow
Cost band	Balanced
Release stage	Stable

Pricing

	Per 1M tokens
Input	$0.20
Output	$0.60

Use this when

Premium background music — Generate high-fidelity instrumental tracks for applications or video projects.
Brand soundtracks — Produce consistent audio assets for marketing campaigns and product demos.
Podcast intros — Create short, polished opening sequences from descriptive text prompts.
Video scoring — Sync custom audio tracks to visual content without external licensing.

Not ideal for

Do not use this model for real-time voice synthesis, interactive audio streaming, or tasks requiring structured JSON outputs.

Pick something else when

You need faster audio generation: use minimax-music.
You need human-like speech synthesis: use minimax-speech-turbo or eleven-turbo-v2-5.
You need the highest production audio quality: use elevenlabs-music.

Example

from openai import OpenAI

client = OpenAI(base_url="https://kymaapi.com/v1", api_key="ky-...")

response = client.chat.completions.create(
    model="minimax-music-pro",
    messages=[{"role": "user", "content": "..."}],
)

Agent query example

Ask the API which models fit, instead of hardcoding an id:

curl "https://kymaapi.com/v1/models?recommended_for=generate-speech&tools=true&quality_tier=frontier-open"

FAQ

Does this model support streaming or real-time audio? No, it generates complete audio files per request and is optimized for quality rather than low-latency streaming. How are repeated prompts billed? Kyma applies prompt caching to repeated input prefixes, billing them at this model’s cached input rate. Can I use this for voice cloning or text-to-speech? No, this endpoint is designed exclusively for music generation. For speech synthesis or voice design, route requests to the dedicated speech or voice models.

​Overview

​Specs

​Pricing

​Use this when

​Not ideal for

​Pick something else when

​Example

​Agent query example

​FAQ