-
### Describe the bug
what happened?
Error caught while starting LLM Server:
environment can only contain strings
### To reproduce
_No response_
### Logs
```shell
Traceback (most recent call l…
-
Remove implicit creation of event loop from `asyncio.get_event_loop` as deprecated from Python 3.12.
### Linked PRs
* gh-126354
-
.
-
### Feature request
Currently, all of the configuration specific for each model is handled via `__init_subclass__` in _configuration.py, and such value will be saved under __openllm___ of the given c…
-
The problem: At the moment, a lot of information related to models is hard coded in the backends.
Think about whether a more general mechanism could help. Like having a single .json file that links…
-
-
Is there a plan to support [Falcon](https://huggingface.co/tiiuae/falcon-7b)
Considering the better performance of Falcon on [OpenLLM leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm…
-
There’s a lot of different LLM deployment providers. How do I easily replace my OpenAI base with their url as a proxy? - https://github.com/petals-infra/chat.petals.dev/issues/20, https://www.banana.d…
-
### Describe the bug
when I do
```python
client.query("What is 3+1?", return_full_text=True)
```
for a running dolly server, it doesn't process `return_full_text` correctly
### To reproduce
_No…
-
### Feature request
Could OpenLLM support MPT models ?
https://www.mosaicml.com/blog/mpt-7b
https://www.mosaicml.com/blog/mpt-30b
### Motivation
Falcon 40B is currently the best Open LLM but have…