Update README.md

README.md

@@ -1,200 +1,90 @@
 ---
 language:
-- en
+  - en
 license: other
 library_name: transformers
 pipeline_tag: text-generation
 tags:
-- python
-- code-generation
-- code-assistant
-- causal-lm
-- full-finetune
-- hunyuan
-- transformers
-- safetensors
-- instruct
+  - gguf
+  - hunyuan
+  - python
+  - code-generation
+  - code-assistant
+  - instruct
+  - conversational
+  - causal-lm
+  - full-finetune
 base_model:
-- tencent/Hunyuan-0.5B-Instruct
-model-index:
-- name: Hunyuan-PythonGOD-0.5B
-  results: []
+  - tencent/Hunyuan-0.5B-Instruct
 datasets:
-- WithinUsAI/Python_GOD_Coder_Omniforge_AI_12k
-- WithinUsAI/Python_GOD_Coder_5k
-- WithinUsAI/Legend_Python_CoderV.1
+  - WithinUsAI/Python_GOD_Coder_Omniforge_AI_12k
+  - WithinUsAI/Python_GOD_Coder_5k
+  - WithinUsAI/Legend_Python_CoderV.1
+model-index:
+  - name: Hunyuan-PythonGOD-0.5B-GGUF
+    results: []
 ---
 
-# Hunyuan-PythonGOD-0.5B
+# Hunyuan-PythonGOD-0.5B-GGUF
 
-Hunyuan-PythonGOD-0.5B is a Python-focused full fine-tune of `tencent/Hunyuan-0.5B-Instruct`, built for code generation, coding assistance, implementation tasks, and instruction-following for Python-heavy workflows.
+**Hunyuan-PythonGOD-0.5B-GGUF** is a compact Python-specialized coding model released in GGUF format for lightweight local inference. It is derived from a full fine-tune of `tencent/Hunyuan-0.5B-Instruct` and is aimed at code generation, Python scripting, debugging help, implementation tasks, and coding-oriented chat workflows.
 
-This release is intended as a compact coding model that keeps the small footprint of the 0.5B Hunyuan base while shifting its behavior toward practical Python generation and code-oriented responses.
+This repo provides quantized GGUF builds for efficient use with llama.cpp-compatible runtimes and other GGUF-serving backends.
 
 ## Model Details
 
-### Model Description
-
-- **Model name:** `gss1147/Hunyuan-PythonGOD-0.5B`
+### Base Model
 - **Base model:** `tencent/Hunyuan-0.5B-Instruct`
-- **Architecture:** causal decoder-only language model
-- **Model family tag:** `hunyuan_v1_dense`
-- **Primary domain:** Python coding / coding assistant
-- **Parameter count:** ~0.5B
-- **Weights format:** safetensors
-- **Tensor type in repo:** F16
-
-### Developed by
-
-- **Shared by:** `gss1147`
-
-### Finetuned from model
-
-- `tencent/Hunyuan-0.5B-Instruct`
-
-## Intended Uses
-
-### Direct Use
-
-This model is intended for:
-
-- Python function generation
-- Python script writing
-- debugging-oriented coding help
-- implementation tasks
-- code completion
-- coding chat assistants
-- lightweight local or cloud inference where a small coding model is preferred
-
-### Downstream Use
-
-Possible downstream uses include:
-
-- code copilots
-- coding bots
-- Python tutoring helpers
-- automation script generation
-- benchmark experimentation for small code LLMs
-
-### Out-of-Scope Use
-
-This model is not designed for:
-
-- safety-critical code deployment without human review
-- medical, legal, or financial decision support
-- secure production code without auditing
-- autonomous execution pipelines without sandboxing
-- guaranteed factual or bug-free code generation
-
-## Training Details
-
-### Training Objective
-
-This model was trained as a **full fine-tune**, not as an adapter-only release.
-
-Based on the training workflow you described and the run logs you shared, this release is meant to represent:
+- **Architecture:** Causal decoder-only language model
+- **Parameter scale:** ~0.5B
+- **Specialization:** Python coding and general code-assistant behavior
+- **Release format:** GGUF
 
-- **full-parameter fine-tuning**
-- **no LoRA**
-- **no QLoRA**
-- **no PEFT adapters in the final model**
-- **standard exported Hugging Face model weights**
+### Included Files
+- `Hunyuan-PythonGOD-0.5B.Q4_K_M.gguf`
+- `Hunyuan-PythonGOD-0.5B.Q5_K_M.gguf`
+- `Hunyuan-PythonGOD-0.5B.f16.gguf`
 
-### Training Data
+## Training Summary
 
-This model was trained on the following datasets:
+This GGUF release is based on a **full fine-tune**, not an adapter-only export.
 
+### Training Datasets
 - `WithinUsAI/Python_GOD_Coder_Omniforge_AI_12k`
 - `WithinUsAI/Python_GOD_Coder_5k`
 - `WithinUsAI/Legend_Python_CoderV.1`
 
-From the training logs you shared, the combined training corpus used:
+### Training Characteristics
+- Full-parameter fine-tuning
+- Python/code-oriented instruction tuning
+- Exported as standard model weights before GGUF conversion
+- Intended for compact coding assistance and local inference experimentation
 
-- **11,760 rows** from `Python_GOD_Coder_Omniforge_AI_12k`
-- **5,000 rows** from `Python_GOD_Coder_5k`
-- **5,000 rows** from `Legend_Python_CoderV.1`
-
-**Total rows:** **21,760**
-
-### Training Procedure
-
-From the training setup you shared, this model was trained with:
-
-- **dual-GPU Kaggle training**
-- **DeepSpeed-assisted distributed training**
-- **full model fine-tuning**
-- **evaluation during training**
-- **final-save upload flow to Hugging Face**
-
-### Sequence Length
-
-- **Practical fine-tuning sequence length:** 4096 tokens
-
-### Context Window Note
-
-If the base model family exposes larger context metadata in config fields, that should not be taken as proof that the full fine-tuning run itself was performed at that larger length. This release should be treated as fine-tuned at **4096 tokens** unless revalidated separately.
-
-## Evaluation
-
-Formal benchmark results are not finalized in this card.
-
-Benchmark attempts were made on free public coding benchmarks such as:
-
-- HumanEval+
-- MBPP+
-- BigCodeBench-style workflows
-
-However, based on the evaluation runs you shared, the harness setup encountered tool/runtime issues during some benchmark attempts, so this card does **not** claim final official benchmark scores yet.
-
-### Observed Training Behavior
-
-From the run logs you shared during training, the model showed:
-
-- strong reduction in training loss over time
-- strong reduction in eval loss over time
-- stable continued learning well into the run
-- increasingly code-specialized behavior relative to the base model
-
-Examples from your shared eval progression included values around:
-
-- ~0.2879 early in training
-- ~0.1071
-- ~0.0604
-- ~0.0550
-- ~0.0422
-- ~0.0329
-- ~0.0266
-- ~0.0299
-- ~0.0290
-
-These are training/eval-run observations, not official public benchmark scores.
-
-## How to Use
+## Intended Uses
 
-### Transformers
+### Good Fits
+- Python function generation
+- Python script writing
+- Debugging assistance
+- Automation script drafting
+- Code-oriented local assistants
+- Small-model coding experiments
 
-```python
-from transformers import AutoTokenizer, AutoModelForCausalLM
-import torch
+### Not Intended For
+- Safety-critical software deployment without review
+- Autonomous execution without sandboxing
+- Guaranteed bug-free or secure code generation
+- Medical, legal, or financial decision support
 
-model_id = "gss1147/Hunyuan-PythonGOD-0.5B"
+## Quantization Notes
 
-tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained(
-    model_id,
-    trust_remote_code=True,
-    torch_dtype=torch.float16,
-    device_map="auto",
-)
+This repo includes multiple tradeoff points:
 
-prompt = "Write a Python function that merges overlapping intervals."
-inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+- **Q4_K_M**: smaller footprint, faster/lighter inference
+- **Q5_K_M**: stronger quality-to-size balance
+- **F16**: highest fidelity in this repo, larger memory cost
 
-with torch.no_grad():
-    outputs = model.generate(
-        **inputs,
-        max_new_tokens=512,
-        do_sample=False,
-    )
+## Example llama.cpp Usage
 
-print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```bash
+./llama-cli -m Hunyuan-PythonGOD-0.5B.Q5_K_M.gguf -p "Write a Python function that validates an email address." -n 256