Overview

ValiantLabs/Llama3.2-3B-Enigma is a 3.2 billion parameter model developed by Valiant Labs, specifically designed for code instruction and general chat. Built on the Llama 3.2 Instruct architecture, it leverages a 32768 token context window to handle complex prompts. The model's training incorporates high-quality synthetic code-instruct data from the sequelbox/Tachibana dataset, alongside generalist synthetic data from sequelbox/Supernova to enhance overall chat capabilities.

Key Capabilities

• High-Quality Code Instruction: Optimized for generating and understanding code-related instructions.
• Llama 3.2 Instruct Format: Utilizes the standard Llama 3.2 Instruct chat format for seamless integration.
• Enhanced General Chat: Supplements code capabilities with strong general conversational performance.

When to Use

This model is ideal for developers and applications requiring a capable and efficient code assistant. Its fine-tuning on specialized datasets makes it particularly effective for tasks involving code generation, explanation, and debugging, while also maintaining robust general chat abilities.