-
**Is your feature request related to a problem? Please describe.**
My understanding is that currently we store a post twice in the database: once as the raw Markdown, and also as the HTML that it is …
-
I'm getting rate-limited very often while testing a low-consumption use case. Doing this with https://github.com/openai/openai-realtime-console doesn't fire the rate limit. It's the same use case and …
-
`timelim_margin` may be a float (because `safety_margin` can be set to, say, `1.2` in `problem.yaml`), it is not rounded to `int` in this case:
https://github.com/Kattis/problemtools/blob/063b68a98…
-
It should be possible to run much faster than what's currently hardcoded. Rate should be adjustable. The limits I see are:
- On the order of 1 us for timings
- Clock rate of 1 MHz
- Internal pro…
-
Update the job spec (hysds-io) page (HySDS github wiki) to highlight a gotcha when it comes to hard/soft time limits.
PGEs can specify a soft and hard time limit, which will cause the worker to tim…
ldang updated
6 years ago
-
A feature can be offered to auto calibrate credit limits for peers as their risk goes down due to good behavior.
ghost updated
7 years ago
-
Cores and ram should be more dynamic, based on the user's privileges, e.g. premium users. Host allocations could still be based on a static algorithm, but there would be an allowance for larger number…
-
-
# Context
File descriptors are an integer representing an abstract indicator used to access files or other input/output resources, such as pipes, network sockets, and devices.
Each process on a…
-
### Describe the issue
CUDA providers failed to build against 12.6 with error error `#221-D`.
### Urgency
_No response_
### Target platform
Windows 11
### Build script
```
./build.bat --config …