Just saw this thread and wanted to jump in. Before I give a full list tho, I gotta ask: what’s your hardware situation and DAW setup? Are you running a dedicated GPU (like an RTX card)? Honestly, it makes a huge difference if you want to run models locally versus paying for cloud credits every month. If you want reliability and zero cost long-term, local is the way to go: • Audiocraft (specifically MusicGen): This is Meta’s open-source model. If you have at least 8GB of VRAM, you can generate high-quality ambient textures locally for free. No subscription traps, and the reliability is 100% since it's on your machine.
• Magenta Studio: This is killer for MIDI generation. It’s a set of free plugins that work inside Ableton or as standalone apps. It uses AI to extend melodies or generate drum patterns without needing a constant internet connection.
• Stable Audio Open: Check this out for sound design. It's a weight-open model that's great for shorter samples and textures, which is perfect for electronic layering. Tbh, cloud-based AI is fine for a bit of fun, but for a serious musician, you don't want your project files tied to a service that might change its pricing or TOS tomorrow (at least that's what worked for me).

Use Suno AI.

This tool is known for its ability to generate full-fledged songs quickly. Users can input a song description, and Suno AI will produce two songs with vocals and background music. It offers a free plan with limited credits and paid plans starting at $8 per month for more extensive use

Gemini AI 2.5 Pro.