Model overview
Pre-quantized GGUF models for acestep.cpp, a portable C++17 implementation of the ACE-Step 1.5 AI Music Generation using GGML. Open http://localhost:8085 in your browser. The embedded WebUI handles everything: write a caption, set lyrics, generate, play, and download tracks. Frozen Qwen3 encoder (28 layers, 1024-dim). The DiT was trained end-to-end with this exact model. Its CondEncoder projection weights (1024 to 2048) are baked into every DiT checkpoint, so the Text-Enc is architecturally locked to 0.6B.
The summary below is compiled from sources by an AI model, not our own testing. Verify selection decisions against original sources and your workload.
Key features
- Pre-quantized GGUF format for efficient CPU inference via acestep.cpp
- Embedded WebUI accessible at http://localhost:8085 for caption input, lyric setting, generation, playback, and download
- Architecturally locked text encoder: frozen Qwen3 (28 layers, 1024-dim) with baked-in CondEncoder projection (1024→2048), tied to 0.6B DiT
Limitations & considerations
- Text encoder is frozen and architecturally locked—cannot be swapped or fine-tuned independently
- Requires acestep.cpp runtime; no native support for other inference backends (e.g., llama.cpp, Ollama)
- No explicit support for multimodal inputs beyond text (e.g., audio conditioning, image prompts)
Source prices & cost analysis
Prices vary by channel, region and configuration. Channel quotes are listed independently and are not treated as a single official provider price.
| Source / channel | Input | Output | Billing unit | Collected at |
|---|---|---|---|---|
| No verified prices yet. Check the source channel. | ||||
Estimate monthly base token costs
Estimates exclude cache discounts, tiered pricing, multimodal surcharges, web search, taxes and additional reasoning output.
Public benchmarks & run configuration
暂无已采集的可核验公开评分。模型未上榜不代表能力较低。
View full indexed rankings →Frequently asked questions
What does 'architecturally locked to 0.6B' mean?
It means the DiT checkpoint embeds fixed projection weights from the frozen Qwen3 encoder (1024-dim) to the DiT's 2048-dim conditioning space—this coupling is baked into the checkpoint and not configurable.
Can I use a different text encoder with this model?
No—the CondEncoder projection weights are permanently embedded in every DiT checkpoint, and the Qwen3 encoder is frozen; swapping encoders is not supported.
Is GPU acceleration supported?
The documentation states acestep.cpp is a portable C++17 implementation using GGML, which primarily targets CPU inference; GPU acceleration is not mentioned and is not guaranteed.
Do open weights permit free commercial use?
Not necessarily. Check the version license, commercial conditions and usage policies. Self-hosting incurs compute and maintenance costs.
Are these real-time billing prices?
These are timestamped source snapshots, not live billing prices. Check the channel for current versions, regions, discounts and billing methods.
Sources & traceability
许可证:mit。模型介绍与第三方评测均以原来源的具体版本和方法为准。