LaaZa
LaaZa
I cannot start stubby 2.6.0 on windows 10. No errors and no output at all. Only -V(version) and -h flags produce any output, others crash. Default configuration. Tested with chocolatey...
### Describe the bug RuntimeError when LLaVA tries to answer to an image. Different errors depending on the settings(clip+projector), but always related to "Half", 16bit. Only working combinations are: *...
### Describe the bug When trying to restart the webui from the "interface mode" tab it gets stuck at the reloading... message and there is no output in the log....
This is a quick implementation of PanQiWei /[AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ) for inference. This is an alternative to the current GPTQ-for-LLaMA hopefully offering a more universally supported option which is not limited to...
Automatically loading models when they are selected is a pain. You can't change the settings before the loading either succeeds or fails and this can take a long time. The...
Adds basic support for MPT models [mosaicml/mpt-7b](https://huggingface.co/mosaicml/mpt-7b) Would solve #69 nice However I think the custom code causes issues and at least I could not quantize above model or use...
Continuation of #364 Mainly brought up to date without changes. Will add test(s) and changes as need. @fxmarty would like input on this, how should we test this? Testing with...
Adds support for [databricks/dbrx](https://huggingface.co/collections/databricks/dbrx-6601c0852a0cdd3c59f71962) **NOT** tested, way too big for me to test and the tiny dummy model [yujiepan/dbrx-tiny-random](https://huggingface.co/yujiepan/dbrx-tiny-random) has too small infeatures. **If possbile please test and report here.**...
This adds support for [InternLM-XComposer2-VL](https://github.com/InternLM/InternLM-XComposer/tree/main#4-bit-model) Based on the code from their repo. Since they use the same `internlm` modeltype but completely different layout, this is implemented in a slightly hacky...