GitLab

- defer closing llm on embedding
- do not override licenses
- remove debugging print line
- reformat model file docs

Fixes #282

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

automatically set num_keep if num_keep < 0

Fixes #297
Fixes #296

Fixes #295

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

missed a backtick

Signed-off-by: Matt Williams <m@technovangelist.com>

First draft of API Docs

configurable rope frequency parameters

Signed-off-by: Matt Williams <m@technovangelist.com>

Signed-off-by: Matt Williams <m@technovangelist.com>

Signed-off-by: Matt Williams <m@technovangelist.com>

langchain JS integration

Signed-off-by: Matt Williams <m@technovangelist.com>

Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>

* resolves: https://github.com/jmorganca/ollama/issues/300 and
https://github.com/jmorganca/ollama/issues/282

* example usage:
```
ollama serve --port 9999 --allowed-origins "http://foo.example.com,http://192.0.0.1"
```

Signed-off-by: Matt Williams <m@technovangelist.com>

Co-authored-by: Michael Yang <mxyng@pm.me>

Co-authored-by: Michael Yang <mxyng@pm.me>