How to use from
Pi
Start the llama.cpp server
# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prithivMLmods/Muscae-Qwen3-UI-Code-4B-GGUF:
Configure the model in Pi
# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:9931/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "prithivMLmods/Muscae-Qwen3-UI-Code-4B-GGUF:"
        }
      ]
    }
  }
}
Run Pi
# Start Pi in your project directory:
pi
Quick Links

Muscae-Qwen3-UI-Code-4B-GGUF

Muscae-Qwen3-UI-Code-4B is a web-UI-focused model fine-tuned on UIGEN-T3-4B-Preview (built upon Qwen3-4B) for controlled Abliterated Reasoning and polished token probabilities, designed exclusively for experimental use. It excels at modern web UI coding tasks, structured component generation, and layout-aware reasoning, making it ideal for frontend developers, UI engineers, and research prototypes exploring structured code generation.

Model Files

File Name Quant Type File Size
Muscae-Qwen3-UI-Code-4B.BF16.gguf BF16 8.05 GB
Muscae-Qwen3-UI-Code-4B.F16.gguf F16 8.05 GB
Muscae-Qwen3-UI-Code-4B.F32.gguf F32 16.1 GB
Muscae-Qwen3-UI-Code-4B.Q2_K.gguf Q2_K 1.67 GB
Muscae-Qwen3-UI-Code-4B.Q3_K_L.gguf Q3_K_L 2.24 GB
Muscae-Qwen3-UI-Code-4B.Q3_K_M.gguf Q3_K_M 2.08 GB
Muscae-Qwen3-UI-Code-4B.Q3_K_S.gguf Q3_K_S 1.89 GB
Muscae-Qwen3-UI-Code-4B.Q4_K_M.gguf Q4_K_M 2.5 GB
Muscae-Qwen3-UI-Code-4B.Q4_K_S.gguf Q4_K_S 2.38 GB
Muscae-Qwen3-UI-Code-4B.Q5_K_M.gguf Q5_K_M 2.89 GB
Muscae-Qwen3-UI-Code-4B.Q5_K_S.gguf Q5_K_S 2.82 GB
Muscae-Qwen3-UI-Code-4B.Q6_K.gguf Q6_K 3.31 GB
Muscae-Qwen3-UI-Code-4B.Q8_0.gguf Q8_0 4.28 GB

Quants Usage

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):

image.png

Downloads last month
423
GGUF
Model size
4B params
Architecture
qwen3
Hardware compatibility
Log In to add your hardware

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

32-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for prithivMLmods/Muscae-Qwen3-UI-Code-4B-GGUF

Finetuned
Qwen/Qwen3-4B
Quantized
(3)
this model

Collection including prithivMLmods/Muscae-Qwen3-UI-Code-4B-GGUF