kihyun-K/kanana-1.5-8b-instruct-2505-Safe-DPO
TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Oct 2, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
The kihyun-K/kanana-1.5-8b-instruct-2505-Safe-DPO is an 8 billion parameter instruction-tuned Llama model developed by kihyun-K. It was fine-tuned using Unsloth and Huggingface's TRL library, enabling 2x faster training. This model is designed for general instruction-following tasks, leveraging its efficient training methodology to provide a capable language model.
Loading preview...
Overview
The kihyun-K/kanana-1.5-8b-instruct-2505-Safe-DPO is an 8 billion parameter instruction-tuned Llama model developed by kihyun-K. This model was fine-tuned using a combination of Unsloth and Huggingface's TRL library, which significantly accelerated its training process, achieving 2x faster fine-tuning.
Key Capabilities
- Instruction Following: Designed to accurately respond to a wide range of user instructions.
- Efficient Training: Benefits from Unsloth's optimizations for faster fine-tuning, making it a potentially resource-efficient option for deployment.
Good for
- General-purpose conversational AI and chatbots.
- Applications requiring a capable instruction-tuned model with an 8 billion parameter footprint.
- Developers looking for models trained with efficient methodologies.