google/gemma-3-4b-it-qat-q4_0-gguf
Image-Text-to-Text • 4B • Updated
• 16.6k • 241
Gemma 3 QAT
Quantization Aware Trained (QAT) Gemma 3 checkpoints. The model preserves similar quality as half precision while using 3x less memory
-
-
google/gemma-3-4b-pt-qat-q4_0-gguf
Image-Text-to-Text • 4B • Updated • 198 • 23 -
google/gemma-3-1b-it-qat-q4_0-gguf
Text Generation • 1.0B • Updated • 1.5k • 115 -
google/gemma-3-1b-pt-qat-q4_0-gguf
Text Generation • 1.0B • Updated • 102 • 12 -
google/gemma-3-12b-it-qat-q4_0-gguf
Image-Text-to-Text • 12B • Updated • 12.1k • 247 -
google/gemma-3-12b-pt-qat-q4_0-gguf
Image-Text-to-Text • 12B • Updated • 296 • 18 -
google/gemma-3-27b-it-qat-q4_0-gguf
Image-Text-to-Text • 27B • Updated • 17.2k • 388
google/gemma-3-27b-pt-qat-q4_0-gguf
Image-Text-to-Text • 27B • Updated • 147 • 29Note ^ GGUFs to be used in llama.cpp and Ollama. We strongly recommend using the IT models.
-
google/gemma-3-4b-it-qat-q4_0-unquantized
Image-Text-to-Text • 4B • Updated • 761 • 10 -
google/gemma-3-27b-it-qat-q4_0-unquantized
Image-Text-to-Text • Updated • 10.6k • 39 -
google/gemma-3-12b-it-qat-q4_0-unquantized
Image-Text-to-Text • Updated • 17k • • 67 -
google/gemma-3-1b-it-qat-q4_0-unquantized
Text Generation • 1.0B • Updated • 270 • 10 -
google/gemma-3-4b-it-qat-int4-unquantized
Image-Text-to-Text • Updated • 385 • 9 -
google/gemma-3-12b-it-qat-int4-unquantized
Image-Text-to-Text • 12B • Updated • 10.8k • 11
google/gemma-3-1b-it-qat-int4-unquantized
Text Generation • 1.0B • Updated • 852 • 13Note ^ unquantized QAT-based checkpoints that allow quantizing while retaining similar quality to half precision