GitLab

Refine error reporting for subprocess crash

Remove no longer supported max vram var

The OLLAMA_MAX_VRAM env var was a temporary workaround for OOM
scenarios.  With Concurrency this was no longer wired up, and the simplistic
value doesn't map to multi-GPU setups.  Users can still set `num_gpu`
to limit memory usage to avoid OOM if we get our predictions wrong.

On windows, the exit status winds up being the search term many
users search for and end up piling in on issues that are unrelated.
This refines the reporting so that if we have a more detailed message
we'll suppress the exit status portion of the message.

Adjust windows ROCm discovery

Refine scheduler unit tests for reliability

Fix context exhaustion integration test for small gpus

The v5 hip library returns unsupported GPUs which wont enumerate at
inference time in the runner so this makes sure we align discovery.  The
gfx906 cards are no longer supported so we shouldn't compile with that
GPU type as it wont enumerate at runtime.

add template validation to modelfile

* distinguish error forwarding

* more coverage

* rm comment

fix parsing tool calls: break on unexpected eofs

* server: only parse tool calls if tools are provided

* still set `resp.Message.Content`

parse tool call as individual objects

remove ToolCall from GenerateResponse

* reopen pr

* tools

* remove tc from stream for now

* ID and Function

* openai expects arguments to be a string (#5739)

* mutually exclusive content and tool calls

* clean up

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

* add suffix

* remove todo

* remove TODO

* add to test

* rm outdated prompt tokens info md

* fix test

* fix test

remove unneeded tool calls

add insert support to generate endpoint