Add GGUF (full + q4_0) and README

README.md

@@ -1,3 +1,26 @@
----
-license: apache-2.0
----
+---
+base_model: microsoft/Phi-3-mini-4k-instruct
+tags:
+- phi3
+- therapy
+- gguf
+- quantized
+library_name: gguf
+license: other
+---
+
+# Therapy Phi-3 Mini — GGUF
+
+This is a fine-tuned version of **Phi-3-mini-4k-instruct**, adapted for supportive, therapeutic-style conversations.
+
+## 📂 Files
+
+- `therapy_phi3.gguf` — Full precision model (~7.6 GB)
+- `therapy_phi3_q4_0.gguf` — 4-bit quantized version (~2–3 GB, faster, lighter)
+
+## 🚀 Usage
+
+### llama.cpp
+```bash
+./llama-cli -m therapy_phi3_q4_0.gguf -p "I'm feeling anxious." -n 100
+

therapy_phi3.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:265dc19a6489ee09dcc942aa7e32b362e6e6f8d3d2099f3f24fc9c8dfee43866
+size 7643296896

therapy_phi3_q4_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee5dc57204011550fc6cbf5fae20cdb9043bac132bb80a3896f4c785d81c8535
+size 2176176768