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
Input Modalities
Output Modalities
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
Data sourced from official provider APIs and documentation
Last updated: Jun 24, 2026
Automatically route workloads to the right model for every task, every time.