LyraNovaHeart/Dazzling-Star-Aurora-32b-v0.0-Experimental-1130

TEXT GENERATIONPricing:Input $2.72 / Output $4.8Concurrent Unit Cost:2Model Size:32.8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Nov 30, 2024Architecture:Transformer0.0K Featherless Exclusive Cold

Dazzling-Star-Aurora-32b-v0.0 is a 32.8 billion parameter experimental language model developed by LyraNovaHeart, built upon the Qwen2.5-32B base. This model is a merge of EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2 and ArliAI/Qwen2.5-32B-ArliAI-RPMax-v1.3, utilizing the TIES merge method. It is designed for general language generation tasks, leveraging its merged components for potentially enhanced conversational and roleplay capabilities within its 32K context window.

Loading preview...

Dazzling-Star-Aurora-32b-v0.0: An Experimental 32B Merge

Dazzling-Star-Aurora-32b-v0.0 is an experimental 32.8 billion parameter language model created by LyraNovaHeart. This model is a result of a midnight merging experiment, aiming to combine the strengths of several Qwen2.5-32B variants.

Key Capabilities & Technical Details

  • Architecture: Based on the Qwen2.5-32B foundation model.
  • Parameter Count: Features 32.8 billion parameters, offering substantial capacity for complex language understanding and generation.
  • Context Length: Supports a context window of 32,768 tokens, enabling processing of lengthy inputs and maintaining coherence over extended conversations.
  • Merge Method: Utilizes the TIES merge method, with Qwen/Qwen2.5-32B serving as the base model.
  • Merged Components: Integrates two distinct models:
    • EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2 (weight: 0.3, density: 0.7)
    • ArliAI/Qwen2.5-32B-ArliAI-RPMax-v1.3 (weight: 0.4, density: 0.8)
  • Instruct Format: Employs the ChatML instruction format for interaction.

Potential Use Cases

This model is suitable for developers and researchers interested in exploring the performance of merged Qwen2.5-32B models. Given its merged components, it may exhibit enhanced capabilities in:

  • General text generation: Creating diverse and coherent text outputs.
  • Conversational AI: Engaging in extended dialogues and maintaining context.
  • Roleplay scenarios: Potentially offering improved performance in character-driven interactions due to the inclusion of an 'RPMax' variant in its merge.

Finetunes
1 models