bartowski releases IQ4_NL quantized GGUF file based on Gemma-4 26B-A4B-it MoE

MeNews · 2026-04-04T04:02:03+00:00

User @outsource_ published the IQ4_NL quantized GGUF file based on the Google Gemma-4 26B-A4B-it MoE model. The model contains approximately 26 billion parameters, and the quantized file size is 14.70 GB, created by bartowski.

MeNews

2026-04-04 04:02:03

Abstract generation in progress

ME News message, April 4 (UTC+8). Recently, user @outsource_ posted a message saying that the IQ4_NL quantized GGUF file based on the Google Gemma-4 26B-A4B-it MoE model has been released. The model has a total of about 26 billion parameters, with about 4 billion parameters activated. The quantized file was made by bartowski, using llama.cpp’s imatrix for revision and quantization. The quantized file name is gemma-4-26B-A4B-it-IQ4_NL.gguf, and its size is 14.70 GB. (Source: InFoQ)

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.