ace-step-1.5 vs sensenova-u1
Two of the top projects in Open models, side by side: who leads on each score part and who each one suits, then setup, license, activity and what each review found.
ace-step-1.5
Text-to-music model with LM planner, DiT decoder, Gradio UI and REST API
sensenova-u1
Open 8B multimodal model for image generation, editing and visual Q&A
Verdict
ace-step-1.5 ranks higher, 73 to 67. Of the 4 score parts we measure for both, ace-step-1.5 leads on 2, sensenova-u1 on 1 and 1 is even.
- Adoptionace-step-1.5 leads by 22 points, 79 to 57: a wide gap.
- FreshnessEven at 100.
- Maintenancesensenova-u1 leads by 6 points, 55 to 49.
- Agent-readyace-step-1.5 leads by 15 points, 45 to 30: a wide gap.
Pick ace-step-1.5 if you are
Musicians and developers generating music locally, with LoRA fine-tuning
- Runs locally with under 4GB VRAM using the DiT-only turbo model; README claims under 10s per song on RTX 3090
- Covers text-to-music, cover, repaint, stem separation, and vocal-to-BGM in one model family
Pick sensenova-u1 if you are
Engineers running open multimodal models for image generation and editing
- One checkpoint covers text-to-image, editing, interleaved generation and VQA
- Native 4K generation and bounding-box or reference-based visual control in U1.5
| What we compare | ace-step-1.5 | sensenova-u1 |
|---|---|---|
| Score parts, out of 100 | ||
| Adoption | 79, popular | 57, popular |
| Freshness | 100, active | 100, active |
| Maintenance | 49, patchy | 55, fair |
| Agent-ready | 45, minimal | 30, minimal |
| Facts from GitHub and the README | ||
| Stars | 15k | 6.9k |
| License | MIT (permissive) | Apache-2.0 (permissive) |
| Last commit | Aug 2026 | Nov 2026 |
| Last release | May 2026 | None published |
| Language | Python | Python |
| Docker | Yes | No |
| GPU | Optional | Not stated |
| arm64 or Apple Silicon | Mentioned | Not stated |
ace-step-1.5
ACE-Step 1.5 generates full songs from text prompts and lyrics using a Qwen3-based language model that plans the song and a diffusion transformer (2B or 4B XL) that renders audio. It also does cover generation, repainting, stem separation, vocal-to-accompaniment and LoRA training. It ships a Gradio UI on port 7860, a REST API on port 8001, a CLI and a Python API.
Who it is for: Musicians and developers generating music locally, with LoRA fine-tuning
Strengths
- Runs locally with under 4GB VRAM using the DiT-only turbo model; README claims under 10s per song on RTX 3090
- Covers text-to-music, cover, repaint, stem separation, and vocal-to-BGM in one model family
- LoRA training from a few songs: 8 songs in about 1 hour on a 12GB 3090, from the Gradio UI
- Several interfaces: Gradio UI, REST API, CLI, Python API, and a separate VST3 plugin
Weaknesses
- XL 4B models need at least 12GB VRAM with offload and quantization, 20GB for comfortable use
- Low-VRAM setups (6GB or less) run without the LM, losing planning and query rewriting
- Quality claims (between Suno v4.5 and v5) are the authors' own, with no numbers in the README
- Python 3.11-3.12 only; CUDA recommended, other backends (MPS, ROCm, XPU, CPU) are secondary
- GPU optional
- Docker + Compose
- Needs uv, Python 3.11-3.12, Gradio
- Models: ACE-Step DiT 2B (base, sft, turbo), ACE-Step XL DiT 4B (base, sft, turbo), acestep-5Hz-lm 0.6B/1.7B/4B (Qwen3-based), LoRA adapters
- port 7860
sensenova-u1
SenseNova-U1 and U1.5 are open-weight native multimodal models (8B-MoT, plus an A3B variant) that handle text-to-image, image editing, interleaved image-text generation and visual question answering in one model. The repo ships Python inference scripts for Transformers, optional GGUF checkpoint loading, 8-step LoRA checkpoints, a ComfyUI workflow and full-parameter fine-tuning code. Production serving is documented separately with LightLLM and LightX2V.
Who it is for: Engineers running open multimodal models for image generation and editing
Strengths
- One checkpoint covers text-to-image, editing, interleaved generation and VQA
- Native 4K generation and bounding-box or reference-based visual control in U1.5
- 8-step LoRA and GGUF checkpoints reduce latency and VRAM use
- Training code, technical reports and a ComfyUI workflow are included
Weaknesses
- README admits dense text, hands, small faces and multi-turn edits can still fail
- Official GGUF weights for U1.5 are not out yet; current GGUFs are community-made
- No tagged release; weights are distributed via Hugging Face and ModelScope
- Needs PyTorch, Transformers, LightLLM (production serving), LightX2V (production serving), gguf and diffusers (optional GGUF)
- Models: SenseNova-U1.5-8B-MoT, SenseNova-U1-8B-MoT, SenseNova-U1-A3B-MoT, safetensors LoRA, GGUF Q8
More in Open models
Building with these?
We help teams pick an open-source AI stack and run it in production. Client work never changes a rank.