Downloads · 30 days
40
2% of all-time downloads
saria-lh/gemma-4-21b-a4b-it-REAP-Q4_K_M-GGUF
gemma-4-21b-a4b-it-REAP-Q4_K_M-GGUF is a machine learning model from saria-lh. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as gemma.
GGUF quantization of 0xSero/gemma-4-21b-a4b-it-REAP.
Downloads · 30 days
40
2% of all-time downloads
All-time downloads
2.2K
Public
Repo size
13.8 GB
Likes
5
Public
Click a slice to open those files.
.gguf13.8 GB · 100%
From the Hugging Face model README
GGUF quantization of 0xSero/gemma-4-21b-a4b-it-REAP.
Router-weighted Expert Activation Pruning — removes the 20% least important MoE experts (25 of 128 → 103 remaining) using calibration-based scoring. Unlike quantization, REAP removes entire experts while keeping the same active parameter count (~4B) per token.
| Property | Value |
|---|---|
| Base Model | google/gemma-4-26b-a4b-it |
| Pruning | REAP 0.20 (103/128 experts) |
| Quantization | Q4_K_M (5.32 BPW) |
| File Size | 12.87 GB |
| Total Parameters | 20.77B |
| Active Parameters | ~4B per token |
| Context Window | 262,144 tokens |
| Architecture | Gemma4 MoE (30 layers) |
Tested on RTX 4070 Ti SUPER (16GB VRAM):
Converted from BF16 safetensors → F16 GGUF using llama.cpp convert_hf_to_gguf.py, then quantized with llama-quantize to Q4_K_M.