jjcoders/suna-v1

👑 JJ coders Suna AI v1

JJ coders Suna AI v1 is a commercial-grade text-to-instrumental music generation foundation model optimized for background music, commercial ads, cinematic scores, and modern electronic beats.

📦 Model Specifications

• Architecture: MusicGen (Transformer Decoders + EnCodec Audio Tokenizer)
• Parameters: 300M (Low-latency streaming tier)
• Audio Output: 32 kHz Studio Audio
• Formats Available:
  • model.safetensors (Native PyTorch / Transformers)
  • JJ_coders_Suna_AI_v1_Q4_K_M.gguf (Quantized GGUF for local CPU/GPU runtimes)

🚀 Quickstart Usage

import torch
import scipy.io.wavfile as wavfile
from transformers import AutoProcessor, MusicgenForConditionalGeneration

model_id = "johanjomet/jj-coders-suna-ai-v1"
processor = AutoProcessor.from_pretrained(model_id)
model = MusicgenForConditionalGeneration.from_pretrained(
    model_id, 
    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32
).to("cuda" if torch.cuda.is_available() else "cpu")

prompt = "an energetic bg music for an apple level ad for JJ coders"
inputs = processor(text=[prompt], padding=True, return_tensors="pt").to(model.device)

with torch.no_grad():
    audio_values = model.generate(**inputs, max_new_tokens=1500)

sampling_rate = model.config.audio_encoder.sampling_rate
wavfile.write("suna_output.wav", sampling_rate, audio_values[0, 0].cpu().numpy())