gradients-io-tournaments/qwen2.5-3b-finetune-upwork1

TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:3.1BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 27, 2026Architecture:Transformer Featherless Exclusive Cold

This is a fine-tuned Qwen2.5-3B-Instruct model, based on the Qwen2ForCausalLM architecture with approximately 3.09 billion parameters and a 32,768 token context length. It is a merged checkpoint optimized for general instruction-following tasks. The model leverages bfloat16 precision for efficient inference.

Loading preview...

Model Overview

This model, gradients-io-tournaments/qwen2.5-3b-finetune-upwork1, is a fine-tuned and merged checkpoint of the Qwen2.5-3B-Instruct base model. It is built upon the Qwen2ForCausalLM architecture, featuring 36 hidden layers, 16 attention heads, and 2 key-value heads.

Key Specifications

  • Parameters: Approximately 3.09 billion parameters.
  • Precision: Utilizes bfloat16 for computational efficiency.
  • Context Length: Supports a substantial context window of 32,768 tokens, enabling processing of longer inputs and generating more coherent responses over extended conversations.
  • Base Model: Derived from Qwen/Qwen2.5-3B-Instruct, indicating its foundation in a robust instruction-tuned language model.

Intended Use Cases

This fine-tuned model is suitable for a variety of applications requiring a capable instruction-following language model within a 3-billion parameter footprint. Its large context window makes it particularly useful for tasks that benefit from extensive contextual understanding, such as:

  • General conversational AI.
  • Text generation and summarization.
  • Question answering based on long documents.
  • Assisting with coding tasks or generating code snippets, given its causal language modeling capabilities.

Finetunes
1 models