AHN GDN For Qwen 2.5 Instruct 3B
ByteDance's AHN GDN For Qwen 2.5 Instruct 3B is a chat model that excels at handling long-context conversations, thanks to its Artificial Hippocampus Networks (AHNs) architecture, which transforms lossless memory into fixed-size compressed representations. This architecture enables the model to efficiently process sequences of up to 32,768 tokens.
Input
Output
Context
33K
Max Output
8K
Parameters
13M
Technical Specifications
Model TypeChat
Context Window32,768 tokens
Max Output Tokens8,192 tokens
Parameters13M
Release DateOct 8, 2025
Training CutoffDec 31, 2023
Licenseother
Open SourceYes
Input Modalities
Text
Output Modalities
Text
Capabilities
Estimated GPU Requirements for AHN GDN For Qwen 2.5 Instruct 3B
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Jun 24, 2026
Start building with the right model.
Automatically route workloads to the right model for every task, every time.