mradermacher

/

model_requests

like 186

It's queued!
You can check for progress at http://hf.tst.eu/status.html or regularly check the model
summary page at https://hf.tst.eu/model#gemma-3-4b-it-qat-q4_0-unquantized-heretic-GGUF for quants to appear.

errors out
it appears the model lacks vision where llama.cpp expects it
(assumption ^)

oh thats sad, i hope that will be fixed or we find a workaround

solved it
add preprocessor_config.json

{
  "image_mean": [0.0, 0.0, 0.0],
  "image_std": [1.0, 1.0, 1.0],
  "image_processor_type": "GemmaImageProcessor"
}

does require you to manually intervene and do that though

also i went ahead and made a quick static q4_0 quant so i guess i won't be needing your quants yet (unless i want imatrix then yeah)

solved it
add preprocessor_config.json

{
  "image_mean": [0.0, 0.0, 0.0],
  "image_std": [1.0, 1.0, 1.0],
  "image_processor_type": "GemmaImageProcessor"
}

does require you to manually intervene and do that though

also i went ahead and made a quick static q4_0 quant so i guess i won't be needing your quants yet (unless i want imatrix then yeah)

Yeah @nicoboss would probably be able to do that or i can duplicate repo