jakeushida/poli-incorrect-untrue

TEXT GENERATIONConcurrency Cost:2Model Size:32.8BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Jun 7, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The jakeushida/poli-incorrect-untrue model is a 32.8 billion parameter Qwen2.5-Coder instruction-tuned causal language model, developed by jakeushida. This model was finetuned from unsloth/qwen2.5-coder-32b-instruct-bnb-4bit, leveraging Unsloth and Huggingface's TRL library for accelerated training. With a 32768 token context length, it is optimized for coding tasks, building upon its Qwen2.5-Coder base.

Loading preview...

Model Overview

The jakeushida/poli-incorrect-untrue model is a 32.8 billion parameter instruction-tuned language model, developed by jakeushida. It is based on the Qwen2.5-Coder architecture and was finetuned from the unsloth/qwen2.5-coder-32b-instruct-bnb-4bit model. This finetuning process utilized Unsloth and Huggingface's TRL library, enabling a 2x faster training speed.

Key Characteristics

  • Architecture: Qwen2.5-Coder, a robust base model known for its coding capabilities.
  • Parameter Count: 32.8 billion parameters, offering a balance of performance and efficiency.
  • Context Length: Supports a substantial context window of 32768 tokens, beneficial for handling longer code snippets or complex instructions.
  • Training Efficiency: Finetuned with Unsloth, which is designed to accelerate the training of large language models.

Intended Use Cases

Given its Qwen2.5-Coder base and instruction-tuned nature, this model is well-suited for:

  • Code Generation: Generating programming code based on natural language prompts.
  • Code Completion: Assisting developers by completing partial code.
  • Code Explanation: Providing explanations for existing code segments.
  • Software Development Tasks: General programming-related tasks where a large context window and coding proficiency are advantageous.