DQN-Labs/qwen306b-rl-exp1

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Sep 21, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

DQN-Labs/qwen306b-rl-exp1 is an 0.8 billion parameter Qwen3 model developed by DQN-Labs. This model was finetuned from unsloth/qwen3-0.6b-base-unsloth-bnb-4bit. It was trained using Unsloth and Huggingface's TRL library, enabling 2x faster training. This model is suitable for applications requiring a compact yet efficient Qwen3-based language model.

Loading preview...

Model Overview

DQN-Labs/qwen306b-rl-exp1 is an 0.8 billion parameter language model developed by DQN-Labs. It is based on the Qwen3 architecture and was finetuned from the unsloth/qwen3-0.6b-base-unsloth-bnb-4bit base model.

Key Training Details

  • Base Model: unsloth/qwen3-0.6b-base-unsloth-bnb-4bit
  • Training Frameworks: This model was finetuned using a combination of Unsloth and Huggingface's TRL library.
  • Training Efficiency: The use of Unsloth enabled a 2x faster training process compared to standard methods.

Potential Use Cases

This model is suitable for developers looking for a compact Qwen3-based model that benefits from efficient training methodologies. Its 0.8 billion parameters make it a good candidate for applications where computational resources are a consideration, while still leveraging the capabilities of the Qwen3 architecture.

Finetunes
3 models