MergekitCloud/mergekit-77

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7BQuant:FP8Context Size:4kTool Calling:SupportedPublished:Aug 31, 2026Architecture:Transformer Featherless Exclusive Cold

MergekitCloud/mergekit-77 is a 7 billion parameter language model created by merging Nexusflow/Starling-LM-7B-beta and FuseAI/FuseChat-7B-VaRM using the SLERP method. This model combines the strengths of its constituent models, offering a versatile foundation for various natural language processing tasks. It is designed to leverage the distinct capabilities of its merged components, providing a balanced performance profile.

Loading preview...

MergekitCloud/mergekit-77: A Merged 7B Language Model

MergekitCloud/mergekit-77 is a 7 billion parameter language model developed by merging two distinct pre-trained models: Nexusflow/Starling-LM-7B-beta and FuseAI/FuseChat-7B-VaRM. This merge was performed using the SLERP (Spherical Linear Interpolation) method, a technique known for smoothly combining model weights.

Merge Details

The model leverages the full 32 layers from both base models, with specific parameter adjustments applied to the self-attention and MLP blocks during the merge process. The base_model for the merge was specified as FuseAI/FuseChat-7B-VaRM.

Key Characteristics

  • 7 Billion Parameters: A moderately sized model suitable for a range of applications.
  • SLERP Merge Method: Utilizes a sophisticated merging technique to blend the capabilities of its components effectively.
  • Combined Strengths: Aims to inherit and combine the best features from both Starling-LM-7B-beta and FuseChat-7B-VaRM, potentially offering improved performance across different benchmarks or use cases compared to individual base models.

Potential Use Cases

This merged model is suitable for general-purpose language tasks, including but not limited to:

  • Text generation
  • Question answering
  • Summarization
  • Chatbot applications

Its merged nature suggests a balanced performance profile, making it a versatile choice for developers looking for a robust 7B model.