anta99/kanana-1.5-8b-instruct-2505-Persona-Merged

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Aug 13, 2026Architecture:Transformer Featherless Exclusive Cold

The anta99/kanana-1.5-8b-instruct-2505-Persona-Merged is an 8 billion parameter instruction-tuned language model with an 8192 token context length. This model is a merged version, indicating a combination of different models or fine-tuning stages. Its primary differentiator is its 'Persona-Merged' characteristic, suggesting an optimization for generating responses with specific personas or conversational styles. It is suitable for use cases requiring nuanced character interaction or role-playing capabilities.

Loading preview...

Model Overview

The anta99/kanana-1.5-8b-instruct-2505-Persona-Merged is an 8 billion parameter instruction-tuned language model, featuring an 8192 token context window. The "Persona-Merged" designation indicates that this model has likely undergone a merging process to enhance its ability to adopt and maintain specific personas during interactions.

Key Characteristics

  • Parameter Count: 8 billion parameters, offering a balance between performance and computational efficiency.
  • Context Length: Supports an 8192 token context, allowing for more extensive and coherent conversations or document processing.
  • Persona-Merged: Optimized for generating responses that adhere to defined personas, making it distinct from general-purpose instruction-tuned models.

Potential Use Cases

  • Role-playing and Conversational AI: Ideal for applications requiring the model to embody specific characters or maintain consistent conversational styles.
  • Interactive Storytelling: Can be used to create dynamic narratives where characters have distinct voices and personalities.
  • Customer Service Bots: Potentially useful for developing bots that need to project a particular brand voice or persona.

Further details regarding its development, training data, and specific performance benchmarks are not provided in the available model card.