Patrick Devine
Patrick Devine
Hey @jackielii , sorry for the slow response. There actually is a _raw mode_ for `/api/generate` if you check out the [API documentation](https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-completion). We've been trying to rethink the way...
Hey @FlorinAndrei sorry this is confusing. `0.5.7` is the current release. `0.5.8` and `0.5.9` are in "pre-release" (you should be able to see this on the releases page); `0.5.8` _would_...
cc @jmorganca
I'm going to go ahead and close this as stale. There have been a lot of improvements w/ embedding models on the ollama engine (i.e. not the legacy llama.cpp engine)...
Hey @bulrush15 , what's wrong with `/bye` or Ctrl-D?
Going to go ahead and close this as stale.
For people who want to try this: * Download the [pre-release of Ollama 0.4.0](https://github.com/ollama/ollama/releases/tag/v0.4.0-rc4) * Pull the vision model with `ollama pull x/llama3.2-vision:11b` (you can find the other quantizations at...
Hey @573932914 , the part that you cut/pasted is just the same as on your screenshot. Can you paste in what `e.response.text` and `e.response.status_code` are set to?
Closing since this is supported now. You can find the tags here: https://ollama.com/library/llama3.1/tags
@emsi and @pacozaa I believe the problem is the uppercase characters in the model name. If you can just change the names to lowercase it _should_ work. We actually have...