pearsonkyle/gemma4-31b-imatrix-mtp-GGUF
Image-Text-to-Text • 31B • Updated • 1.53k • 3
Gemma
Quantized Gemma 4 models, including the 31B imatrix MTP GGUF.
-
pearsonkyle/gemma4-e4b-w4a16-mtp-vLLM
Text Generation • UpdatedNote W4A16 (full lm_head) + MTP drafter, ~130 tok/s single-GPU vLLM, tool-calling verified