Dohyeon1/ERNIE-HC-SMoE-ngroups56

TEXT GENERATIONPricing:Input $0.32 / Cached $0.016 / Output $1.6Concurrent Unit Cost:1Model Size:21BQuant:FP8Context Size:32kPublished:Sep 19, 2026Architecture:Transformer Featherless Exclusive Cold

Dohyeon1/ERNIE-HC-SMoE-ngroups56 is a 21 billion parameter language model with a context length of 32768 tokens. This model is based on the Mixture-of-Experts (MoE) architecture, indicated by 'SMoE-ngroups56', suggesting it leverages multiple expert networks for improved efficiency or performance. While specific training details and primary differentiators are not provided in the available information, its large parameter count and context window suggest it is designed for complex language understanding and generation tasks.

Loading preview...

Model Overview

Dohyeon1/ERNIE-HC-SMoE-ngroups56 is a substantial language model featuring 21 billion parameters and a context length of 32768 tokens. The 'SMoE-ngroups56' in its name indicates a Sparse Mixture-of-Experts architecture, which typically allows models to scale to a large number of parameters while maintaining computational efficiency during inference by activating only a subset of 'expert' networks for any given input. This design often leads to strong performance across a variety of tasks.

Key Characteristics

  • Architecture: Sparse Mixture-of-Experts (SMoE) with 56 groups, suggesting a focus on efficient scaling and potentially specialized processing.
  • Parameter Count: 21 billion parameters, placing it in the category of large language models capable of advanced reasoning and generation.
  • Context Length: A significant context window of 32768 tokens, enabling the model to process and understand very long inputs, which is beneficial for tasks requiring extensive context such as document summarization, long-form content generation, or complex code analysis.

Potential Use Cases

Given its architecture and scale, this model is likely suitable for:

  • Advanced Text Generation: Creating detailed articles, stories, or code snippets.
  • Complex Question Answering: Answering questions that require understanding large documents or multiple pieces of information.
  • Long-form Summarization: Condensing extensive texts while retaining key information.
  • Code Understanding and Generation: Its large context window could be particularly useful for handling large codebases or complex programming tasks.

Finetunes
1 models