g-assismoraes/DeltaP2S-Llama2-13B-P2S-CodeLlama7B

TEXT GENERATIONPricing:Input $1.5 / Output $2.1Concurrent Unit Cost:1Model Size:13BQuant:FP8Context Size:4kPublished:Aug 24, 2026Architecture:Transformer Featherless Exclusive Cold

DeltaP2S-Llama2-13B-P2S-CodeLlama7B is a 13 billion parameter language model developed by g-assismoraes, built upon the Llama 2 architecture and fine-tuned with CodeLlama 7B. This merged checkpoint is the result of a standalone Delta-P2S experiment, indicating a focus on specific performance improvements or adaptations. It is designed for tasks benefiting from its specialized training, likely in code-related applications given its CodeLlama integration.

Loading preview...

Model Overview

DeltaP2S-Llama2-13B-P2S-CodeLlama7B is a 13 billion parameter model developed by g-assismoraes. This model represents a merged checkpoint from a standalone Delta-P2S experiment, combining elements from the Llama 2 architecture with fine-tuning derived from CodeLlama 7B.

Key Characteristics

  • Architecture: Based on the Llama 2 family, enhanced with CodeLlama 7B integration.
  • Parameter Count: 13 billion parameters, offering a balance between capability and computational requirements.
  • Origin: A merged checkpoint from a specific "Delta-P2S" experimental training run.
  • Context Length: Supports a context length of 4096 tokens.

Potential Use Cases

Given its foundation and the integration of CodeLlama, this model is likely optimized for:

  • Code Generation: Assisting with writing and completing code snippets.
  • Code Understanding: Analyzing and explaining programming logic.
  • Software Development Tasks: Potentially useful for debugging, refactoring, or generating documentation related to code.

This model is a product of a targeted training experiment, suggesting specialized performance in its intended domain.