yashsoni78
/

mcp_tool_model

@@ -1,202 +1,86 @@
----
-base_model: mistralai/Mistral-7B-Instruct-v0.2
-library_name: peft
----
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-### Model Sources [optional]
-<!-- Provide the basic links for the model. -->
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-## Uses
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-### Direct Use
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-[More Information Needed]
-### Downstream Use [optional]
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-[More Information Needed]
-### Out-of-Scope Use
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-[More Information Needed]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### Recommendations
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-## How to Get Started with the Model
-Use the code below to get started with the model.
-[More Information Needed]
 ## Training Details
-### Training Data
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-[More Information Needed]
-### Training Procedure
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-#### Preprocessing [optional]
-[More Information Needed]
-#### Training Hyperparameters
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-#### Speeds, Sizes, Times [optional]
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-[More Information Needed]
-## Evaluation
-<!-- This section describes the evaluation protocols and provides the results. -->
-### Testing Data, Factors & Metrics
-#### Testing Data
-<!-- This should link to a Dataset Card if possible. -->
-[More Information Needed]
-#### Factors
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-[More Information Needed]
-#### Metrics
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-[More Information Needed]
-### Results
-[More Information Needed]
-#### Summary
-## Model Examination [optional]
-<!-- Relevant interpretability work for the model goes here -->
-[More Information Needed]
-## Environmental Impact
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-## Technical Specifications [optional]
-### Model Architecture and Objective
-[More Information Needed]
-### Compute Infrastructure
-[More Information Needed]
-#### Hardware
-[More Information Needed]
-#### Software
-[More Information Needed]
-## Citation [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]
-### Framework versions
-- PEFT 0.15.2

+# MCP Tool-Calling Agent (v1)
+This repository contains a specialized version of `mistralai/Mistral-7B-Instruct-v0.2`, fine-tuned to function as a reasoning engine for a tool-calling AI agent.
+The model has been trained to understand natural language requests related to a custom set of "MCP" tools and translate them into a specific, structured format suitable for execution in an application backend.
+---
+## Model Description
+This model was fine-tuned using **Parameter-Efficient Fine-Tuning (PEFT)** with **LoRA** on a custom, high-quality dataset. Its primary skill is to receive a user prompt and generate a tool call in a specific, sandboxed format. While the fine-tuning has exposed it to several tool types, its core capability is understanding the intent to use a tool and structuring the output accordingly.
+The model expects prompts in a `SYSTEM-USER-ASSISTANT` format and has been trained to generate tool calls with the 'tool_code' format..
+## Intended Use
+This model is not a general-purpose chatbot. It is a specialized component intended to be used within a larger application or "agent" that can parse and execute the generated code.
+* **Primary Use:** Translating natural language commands into structured code for automation.
+* **Out of Scope:** General conversation, creative writing, or tasks outside its trained toolset.
+---
+## How to Use
+The model is a PEFT adapter, meaning you must load it on top of the original base model. The following code provides a complete example of how to load the model from the Hub and run inference.
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from peft import PeftModel
+# --- 1. Configuration ---
+BASE_MODEL_REPO_ID = "mistralai/Mistral-7B-Instruct-v0.2"
+# Replace with your actual model repository ID on the Hub
+FINE_TUNED_REPO_ID = "yashsoni78/mcp_tool_model"
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+# --- 2. Load Model and Tokenizer ---
+print(f"Loading base model: {BASE_MODEL_REPO_ID}")
+base_model = AutoModelForCausalLM.from_pretrained(
+    BASE_MODEL_REPO_ID,
+    torch_dtype=torch.bfloat16,
+    device_map=DEVICE
+)
+print(f"Loading fine-tuned adapter & tokenizer from: {FINE_TUNED_REPO_ID}")
+tokenizer = AutoTokenizer.from_pretrained(FINE_TUNED_REPO_ID)
+# Resize token embeddings to account for any special tokens added during training
+base_model.resize_token_embeddings(len(tokenizer))
+# Load the LoRA adapter and merge it into the base model for faster inference
+model = PeftModel.from_pretrained(base_model, FINE_TUNED_REPO_ID)
+model = model.merge_and_unload()
+model.eval() # Set the model to evaluation mode
+print("✅ Model loaded successfully.")
+# --- 3. Run Inference ---
+system_prompt = "You are an expert assistant that uses MCP tools. When a tool is required, you must respond *only* with the 'tool_code' format."
+user_prompt = "What's the status of my 'database-main' VM?"
+formatted_prompt = f"SYSTEM: {system_prompt}\nUSER: {user_prompt}\nASSISTANT:"
+inputs = tokenizer(formatted_prompt, return_tensors="pt").to(DEVICE)
+outputs = model.generate(**inputs, max_new_tokens=150, pad_token_id=tokenizer.eos_token_id)
+response = tokenizer.decode(outputs[0], skip_special_tokens=True).split("ASSISTANT:")[1].strip()
+print("\n--- Model Output ---")
+print(response)
+````
 ## Training Details
+  * **Base Model:** `mistralai/Mistral-7B-Instruct-v0.2`
+  * **Fine-Tuning Method:** PEFT (LoRA) with 4-bit quantization (`bitsandbytes`).
+  * **Dataset:** The model was trained on a curated, high-quality dataset containing a balanced mix of tool-calling examples and conversational ("negative") examples to teach it when *not* to call a tool. Special tokens `[TOOL_CODE_START]` and `[TOOL_CODE_END]` were added to the vocabulary.
+  * **Training Procedure:** The model was trained using the `SFTTrainer` from the TRL library.
+## Limitations and Bias
+  * **Syntax Hallucination:** As a probabilistic model, it may occasionally generate tool calls with slightly incorrect syntax (e.g., wrong object names, extra parameters). It is **highly recommended** to use this model within an application that performs a `Parse -> Validate -> Execute` loop to ensure safety and reliability.
+  * **Scope:** The model's knowledge is limited to the patterns and tools seen during fine-tuning. It cannot use new tools without being re-trained.
+  * **Language:** The model was trained exclusively on English language prompts.
+```