-
47)
[//]: # (REMBEMBER! Add links to things related to the bug using for example:)
[//]: # (https://www.wowhead.com/cata)
**Description:**
Quest does not work. After accepting and collecting a…
-
### Feature request
I have been using [CANINE](https://arxiv.org/pdf/2103.06874) for my experiments and I see that there does not exist a fast version of the tokenizer for the model. CANINE accepts u…
-
works for normal users, but wont suggest friends once typing in part of their username
/
![image](https://github.com/user-attachments/assets/0b3d8177-ca38-4609-823f-93981d650f57)
![image](https:/…
-
Right now we render every 50ms, but in reality we need to go as fast as possible. We can send frames endlessly, but they should do minimal work unless something has changed. And they absolutely should…
-
As a developer, I want code changes to reflect immediately during development.
-
```py
import langroid as lr
from langroid.language_models import Role, LLMMessage
cfg = lr.language_models.OpenAIGPTConfig(
chat_model = lr.language_models.OpenAIChatModel.GPT4
)
mdl = lr.…
-
Since #23, there have been many implementations in Node using fast APIs, and I want to open this issue to encourage to revisit them and probably take a look at other places where we could implement th…
-
Currently, both the old ASDF erlang plugin and the built-in erlang plugin depend on kerl to build the erlang binary every single time.
This is acceptable on a dev machine, but when trying to use in…
-
Based on previous discussions and benchmarking, we plan to roll this out in two phases:
A note from me:
```
I believe the best path forward is:
Soon:
Decide on the algorithm (I think it's p…
-
Ever anxious when working on a new codebase. Always second guessing when faced with a new feature request. Here’s a list of techniques and tools that will help you draft airtight PR’s in fractions of …