TareksGraveyard/Primogenitor-V2-LLaMa-70B

TEXT GENERATIONConcurrency Cost:4Model Size:70BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Feb 22, 2025License:llama3.3Architecture:Transformer0.0K Gated Cold

TareksGraveyard/Primogenitor-V2-LLaMa-70B is a 70 billion parameter language model based on the LLaMa architecture, created by TareksGraveyard. This model is a merge of several pre-trained LLaMa-based models, including Wayfarer, using the SCE merge method. It is designed to combine the strengths of its constituent models, offering a versatile foundation for various generative AI tasks with a 32768 token context length.

Loading preview...

Primogenitor V2: A Merged LLaMa-Based Model

Primogenitor V2 is a 70 billion parameter language model developed by TareksGraveyard, built upon the LLaMa architecture. It distinguishes itself as a merged model, combining the capabilities of multiple pre-trained LLaMa-based models using the SCE (Selective Channel Expansion) merge method. This approach aims to integrate diverse strengths from its components into a single, cohesive model.

Key Merge Details

  • Base Model: The merging process utilized nbeerbower/Llama-3.1-Nemotron-lorablated-70B as its foundational base.
  • Constituent Models: Primogenitor V2 incorporates contributions from a variety of LLaMa-based models, including:
    • Sao10K/L3.1-70B-Hanami-x1
    • Sao10K/70B-L3.3-Cirrus-x1
    • LatitudeGames/Wayfarer-Large-70B-Llama-3.3
    • SicariusSicariiStuff/Negative_LLAMA_70B
    • TheDrummer/Anubis-70B-v1
    • EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1

Purpose and Potential Use Cases

By merging these distinct models, Primogenitor V2 is designed to inherit and synthesize their individual strengths, potentially offering enhanced performance across a broader range of generative tasks. Its 70B parameter count and 32768 token context length make it suitable for complex applications requiring extensive context understanding and generation.