Point Space to org-owned interpreter model

README.md

@@ -27,7 +27,7 @@ Site-invariant Autism Spectrum Disorder classification from resting-state fMRI,
 | Model | Description |
 |---|---|
 | [Yatsuiii/brain-connect-gcn](https://huggingface.co/Yatsuiii/brain-connect-gcn) | Adversarial Brain Mode Network — 20-model LOSO GCN ensemble, AUC 0.7298 cross-site |
-| [Yatsuiii/asd-interpreter-merged](https://huggingface.co/Yatsuiii/asd-interpreter-merged) | Qwen2.5-7B fine-tuned clinical interpreter — generates natural language reports from saliency scores |
+| [lablab-ai-amd-developer-hackathon/asd-interpreter-merged](https://huggingface.co/lablab-ai-amd-developer-hackathon/asd-interpreter-merged) | Qwen2.5-7B fine-tuned clinical interpreter — generates natural language reports from saliency scores |
 
 ## Pipeline
 

app.py

@@ -154,7 +154,7 @@ def preprocess(bold):
 # ── LLM (Qwen2.5-7B fine-tuned on AMD MI300X, served via vLLM on MI300X) ───
 
 _VLLM_URL   = os.environ.get("VLLM_BASE_URL", "")
-_LLM_MODEL  = "Yatsuiii/asd-interpreter-merged"
+_LLM_MODEL  = "lablab-ai-amd-developer-hackathon/asd-interpreter-merged"
 _HF_TOKEN   = os.environ.get("HF_TOKEN", "")
 
 # Pre-generated reports for demo subjects (instant display, no LLM latency)