Josh Leverette

I would love for this option to exist just because Whisper is so much more accurate.

@phyzical out of curiosity, which whisper container do you use (to be clear, I have not contributed to open-webui, but I am curious about a whisper server)

I do not understand why this API is using a very non-standard "X-Payload" header to contain the request body, and despite the link to `/docs`, there is no real documentation...

@sidroopdaska With a good enough synthesized voice, I would enjoy being able to paste in an article and have it read it to me sometimes. So, I was just playing...

I personally ran into this issue this morning. Stalebots are counterproductive. This is a very real issue, as confirmed by multiple people in this thread. The lack of response from...

@TheJoeFin I’ve never once seen any program other than PowerToys fail to copy something to the clipboard on Windows, which makes me skeptical that the problem truly is with Windows,...

On a related note, even when using 2K context size, the 3-bit model never offloads all 33 layers to the GPU, even though I know it works fine with all...

@jmorganca Here, I have uploaded the last 4000 lines of log output. The end of the log is the most relevant. [ollama.txt](https://github.com/jmorganca/ollama/files/13894792/ollama.txt)

@IAMBUDE I had tried that, but it no longer works: https://github.com/jmorganca/ollama/issues/1906 I don’t want to manage the layer offload count anyways. It’s very hard to get that number right, especially...

@jmorganca Unfortunately, as I mentioned at the end of the Zero Layers offload issue a few hours ago, I can still reproduce this OOM consistently on 0.1.20. I can try...