GitLab

- add the optional `raw` generate request parameter to bypass prompt formatting and response context
-add raw request to docs

- only reload the running llm if the model has changed, or the options for loading the running model have changed
- rename loaded llm to runner to differentiate from loaded model image
- remove logic which keeps the first system prompt in the generation context

- remove new lines from llama.cpp error messages relayed to client
- check api option types and return error on wrong type
- change num layers from 95% VRAM to 92% VRAM

* update streaming request accept header
* add optional stream param to request bodies

* include seed in params for llama.cpp server and remove empty filter for temp

* relay default predict options to llama.cpp

- reorganize options to match predict request for readability

* omit empty stop

---------
Co-authored-by: hallh <hallh@users.noreply.github.com>

* linux gpu support
* handle multiple gpus
* add cuda docker image (#488)
---------
Co-authored-by: Michael Yang <mxyng@pm.me>

* remove c code
* pack llama.cpp
* use request context for llama_cpp
* let llama_cpp decide the number of threads to use
* stop llama runner when app stops
* remove sample count and duration metrics
* use go generate to get libraries
* tmp dir for running llm

Co-Authored-By: Jeffrey Morgan <jmorganca@gmail.com>

- default to embeddings enabled
- move embedding logic for loaded model to request
- allow embedding full directory
- close llm on reload

Fixes #297
Fixes #296

num_keep defines how many tokens to keep in the context when truncating
inputs. if left to its default value of -1, the server will calculate
num_keep to be the left of the system instructions

- read runner options from map to see what was specified explicitly and overwrite zero values