Update README.md

README.md

@@ -1,21 +1,117 @@
 ---
-base_model: dipta007/GanitLLM-0.6B-SFT-1404
-tags:
-- text-generation-inference
-- transformers
-- unsloth
-- qwen3
+library_name: transformers
 license: apache-2.0
+base_model: Qwen/Qwen3-0.6B
+pipeline_tag: text-generation
 language:
-- en
+  - bn
+  - en
+tags:
+  - math
+  - bengali
+  - reasoning
+  - grpo
+datasets:
+  - dipta007/Ganit
 ---
 
-# Uploaded finetuned  model
+# GanitLLM-0.6B_SFT_GRPO
+
+[![Paper](https://img.shields.io/badge/arXiv-Paper-red)](https://arxiv.org/)
+[![Dataset](https://img.shields.io/badge/HuggingFace-Dataset-yellow)](https://huggingface.co/datasets/dipta007/Ganit)
+[![Models](https://img.shields.io/badge/HuggingFace-Models-orange)](https://huggingface.co/collections/dipta007/ganitllm)
+
+## Highlights
+
+**GanitLLM-0.6B_SFT_GRPO** is our smallest Bengali mathematical reasoning model trained with SFT followed by standard GRPO. Ideal for resource-constrained deployments. Key improvements over the base Qwen3-0.6B model:
+
+- **+24.0 accuracy** on Bn-MGSM benchmark (8.4 → 32.4)
+- **+40.3 accuracy** on Bn-MSVAMP benchmark (12.2 → 52.5)
+- **88.45% Bengali reasoning** (vs 12.43% for base model)
+- **80.6% fewer tokens** in generated solutions (1265 → 246 words)
+
+## Model Overview
+
+| Property | Value |
+|----------|-------|
+| **Model Type** | Causal Language Model |
+| **Base Model** | Qwen/Qwen3-0.6B |
+| **Parameters** | 0.6B |
+| **Training** | SFT + GRPO |
+| **Context Length** | 4,096 tokens |
+| **Language** | Bengali, English |
+
+## Training Details
+
+This model was trained using a two-stage pipeline:
+
+1. **Supervised Fine-Tuning (SFT)**: Trained on GANIT-SFT (~11k examples) to ground reasoning in Bengali
+2. **GRPO**: Standard reinforcement learning with random sampling on GANIT-RLVR (~7.3k examples)
+
+### Reward Functions
+- **Format Reward**: Validates `<think>` and `<answer>` tag structure
+- **Correctness Reward**: +2.0 for Bengali answer match, +1.0 for English match
+- **Bengali Reasoning Reward**: Ensures >80% Bengali text in reasoning
+
+## Quickstart
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_name = "dipta007/GanitLLM-0.6B_SFT_GRPO"
+
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto",
+    device_map="auto"
+)
+
+problem = "একটি দোকানে ১২টি আপেল আছে। যদি ৫টি আপেল বিক্রি হয়, তাহলে কতটি আপেল বাকি থাকবে?"
+
+prompt = f"""A conversation takes place between the user and the assistant. The user asks a question, and the assistant solves the problem. Please reason step by step in Bengali, and put your final answer in the <answer> </answer> tags.
+
+Question: {problem}"""
+
+messages = [{"role": "user", "content": prompt}]
+text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+
+generated_ids = model.generate(**model_inputs, max_new_tokens=2048, temperature=0.7)
+output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
+response = tokenizer.decode(output_ids, skip_special_tokens=True)
+print(response)
+```
+
+### Using vLLM
+
+```bash
+vllm serve dipta007/GanitLLM-0.6B_SFT_GRPO --max-model-len 4096
+```
+
+## Performance
+
+| Model | Bn-MGSM | Bn-MSVAMP | Avg. Words | Bengali % |
+|-------|---------|-----------|------------|-----------|
+| Qwen3-0.6B (base) | 8.40 | 12.20 | 1265 | 12.43% |
+| **GanitLLM-0.6B_SFT_GRPO** | **32.40** | **52.50** | **246** | **88.45%** |
+
+## Related Models
+
+| Model | Parameters | Training | Link |
+|-------|------------|----------|------|
+| GanitLLM-4B_SFT_GRPO | 4B | SFT + GRPO | [Link](https://huggingface.co/dipta007/GanitLLM-4B_SFT_GRPO) |
+| GanitLLM-1.7B_SFT_GRPO | 1.7B | SFT + GRPO | [Link](https://huggingface.co/dipta007/GanitLLM-1.7B_SFT_GRPO) |
+| GanitLLM-0.6B_SFT_CGRPO | 0.6B | SFT + CGRPO | [Link](https://huggingface.co/dipta007/GanitLLM-0.6B_SFT_CGRPO) |
+| **GanitLLM-0.6B_SFT_GRPO** | 0.6B | SFT + GRPO | [Link](https://huggingface.co/dipta007/GanitLLM-0.6B_SFT_GRPO) |
+| GanitLLM-0.6B_CGRPO | 0.6B | CGRPO | [Link](https://huggingface.co/dipta007/GanitLLM-0.6B_CGRPO) |
+
+## Citation
 
-- **Developed by:** dipta007
-- **License:** apache-2.0
-- **Finetuned from model :** dipta007/GanitLLM-0.6B-SFT-1404
+```bibtex
+will be updated
+```
 
-This qwen3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
+## License
 
-[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
+This model is released under the Apache 2.0 License.