Update README.md

README.md

@@ -1,21 +1,111 @@
 ---
-base_model: unsloth/LFM2.5-1.2B-Instruct
+license: apache-2.0
+base_model: LiquidAI/LFM2.5-1.2B-Instruct
 tags:
-- text-generation-inference
-- transformers
+- function-calling
+- liquid-neural-network
 - unsloth
-- lfm2
-license: apache-2.0
+- tool-use
+- gguf
+- conversational
+datasets:
+- NovachronoAI/Nova-Synapse-Function-Calling
+library_name: transformers
+pipeline_tag: text-generation
 language:
 - en
 ---
 
-# Uploaded finetuned  model
+# 🌊 LFM 2.5 1.2B - Nova Synapse (Function Calling)
+
+<div align="center">
+
+![Unsloth Fine-tuning](https://img.shields.io/badge/Fine--Tuned%20with-Unsloth-blue?style=for-the-badge)
+![Liquid AI](https://img.shields.io/badge/Architecture-Liquid%20Neural%20Network-cyan?style=for-the-badge)
+![Function Calling](https://img.shields.io/badge/Task-SOTA%20Function%20Calling-orange?style=for-the-badge)
+![Size](https://img.shields.io/badge/Params-1.2B-green?style=for-the-badge)
+
+</div>
+
+## 🚀 Model Overview
+**LFM2.5-1.2B-Nova-Function-Calling** is a specialized fine-tune of Liquid AI's revolutionary **Liquid Neural Network (LFM 2.5)**. Despite its small size (1.2B parameters), this model rivals 7B+ class models in specific tasks due to its hybrid architecture.
+
+This model has been **specifically engineered for robust Function Calling**, allowing it to seamlessly convert natural language user queries into structured JSON inputs for tools, APIs, and software agents.
+
+### 🌟 Key Features
+* **Hyper-Efficient:** Runs on extremely low-resource hardware (phones, Raspberry Pi, older laptops) thanks to the 1.2B Liquid architecture.
+* **Precision Tuned:** Achieved a training loss of **2.63**, mastering structured JSON syntax without overfitting.
+* **ChatML Native:** Uses the standard `<|im_start|>` format for easy integration.
+* **GGUF Ready:** Available in all quantization levels (from 16-bit down to 2-bit).
+
+---
+
+## 📚 Dataset
+This model was trained on **[NovachronoAI/Nova-Synapse-Function-Calling](https://huggingface.co/datasets/NovachronoAI/Nova-Synapse-Function-Calling)**.
+
+* **Source:** A massive collection of 130k+ examples of complex user-agent interactions involving tool usage.
+* **Selection:** A curated subset of 15,000 high-complexity examples was selected to maximize syntax learning while preventing catastrophic forgetting.
+* **Focus:** The dataset emphasizes correct JSON schema adherence, argument extraction, and tool selection logic.
+
+---
+
+## 💻 Quick Start (Inference)
+
+### 1. Using Transformers
+You need the latest `transformers` and `unsloth` libraries to run Liquid architectures.
+
+```python
+from unsloth import FastLanguageModel
+import torch
+
+# Load the model
+model, tokenizer = FastLanguageModel.from_pretrained(
+    model_name = "NovachronoAI/LFM2.5-1.2B-Nova-Function-Calling-Full", # or use the GGUF repo
+    max_seq_length = 4096,
+    dtype = None,
+    load_in_4bit = True,
+)
+FastLanguageModel.for_inference(model)
+
+# Define the Prompt (ChatML Format)
+prompt = """<|im_start|>user
+I need to calculate the area of a circle with a radius of 5.
+<|im_end|>
+<|im_start|>assistant
+"""
 
-- **Developed by:** NovachronoAI
-- **License:** apache-2.0
-- **Finetuned from model :** unsloth/LFM2.5-1.2B-Instruct
+# Generate
+inputs = tokenizer([prompt], return_tensors = "pt").to("cuda")
+outputs = model.generate(**inputs, max_new_tokens = 128, use_cache = True)
+print(tokenizer.batch_decode(outputs)[0].split("<|im_start|>assistant")[-1])
 
-This lfm2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
+Expected Output:
+<tool_call>
+{"name": "calculate_circle_area", "arguments": {"radius": 5}}
+</tool_call>
 
-[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
+2. Using GGUF (llama.cpp / Ollama)
+This model is available in GGUF format in the companion repository: NovachronoAI/LFM2.5-1.2B-Nova-Function-Calling-GGUF
+ * Recommended: q4_k_m.gguf (Balanced Speed/Quality - ~800MB)
+ * Max Quality: f16.gguf (Lossless - ~2.5GB)
+ * Max Speed: q2_k.gguf (Extreme Speed - ~400MB)
+⚙️ Training Details
+| Parameter | Value |
+|---|---|
+| Base Model | LiquidAI/LFM2.5-1.2B-Instruct |
+| Framework | Unsloth + Hugging Face TRL |
+| Hardware | NVIDIA Tesla T4 (Kaggle) |
+| Epochs | ~2 (600 Steps) |
+| Learning Rate | 2e-4 |
+| Scheduler | Linear |
+| Quantization | 4-bit (QLoRA) |
+Training Trajectory
+The model showed rapid adaptation to the JSON syntax, dropping from a random-guess loss of 11.6 to a highly capable 2.63.
+ * Start: Loss 11.68 (Step 10)
+ * Convergence: Loss ~3.0 (Step 160)
+ * Final: Loss 2.63 (Step 600)
+📜 License
+This model is fine-tuned from LiquidAI/LFM2.5-1.2B-Instruct. Please refer to the original Liquid AI license terms for commercial use. The fine-tuning dataset and adapters are released under Apache 2.0.
+<div align="center">
+Built with ❤️ by <b>NovachronoAI</b> using <a href="https://github.com/unslothai/unsloth">Unsloth</a>
+</div>