joyfox/JoyFox-Reverie-27B

VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Oct 14, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

JoyFox-Reverie-27B is a 27 billion parameter conversational language model developed by JoyFox, derived from Qwen3.8-27B, with a context length of 32768 tokens. It is specifically fine-tuned for character roleplay, companion conversations, and shared storytelling, allowing users to define personas and narrative styles. The model retains visual components for image and video inputs, making it suitable for multimodal conversational applications.

Loading preview...

JoyFox-Reverie-27B: Conversational AI for Imagined Worlds

JoyFox-Reverie-27B is a 27 billion parameter conversational model developed by JoyFox, built upon the Qwen3.8-27B architecture. It is specifically designed to excel in interactive and imaginative dialogue scenarios, focusing on character roleplay, companion conversations, and collaborative storytelling. The model supports a substantial context length of 32768 tokens, enabling extended and coherent interactions.

Key Capabilities

  • Character Roleplay: Users can define detailed personas, including background, voice, relationships, and scene instructions, for the model to embody.
  • Companion Conversations: Facilitates natural and friendly dialogue for everyday topics and personal reflection, adapting to a conversational style.
  • Shared Storytelling: Enables collaborative narrative creation, allowing participants to define which characters each controls and build scenes together.
  • Flexible Expression: Offers controls for specifying reply language, narrative perspective, response length, and the balance between action and dialogue.
  • Visual Context: Retains upstream visual components, supporting image and video inputs through compatible processors and runtimes.
  • Thinking Controls: Provides options to enable or disable internal reasoning processes, with adjustable effort levels (low, medium, xhigh) and the ability to preserve reasoning content.

Good For

  • Interactive Fiction and Gaming: Creating dynamic character interactions and branching narratives.
  • Virtual Companions and Chatbots: Developing engaging and personalized conversational agents.
  • Creative Writing Assistance: Collaborating on stories, scripts, and dialogue generation.
  • Multimodal Conversational AI: Applications requiring both text and image/video understanding within a dialogue context.

Finetunes
1 models