-
## Summary
Piexif corrupts the `VP8X` flags for animated WebP data.
## Steps to Reproduce
The following commit on my fork adds two animated webp test files: https://github.com/hMatoba/Piexif/comm…
-
I am using BatchedInferencePipeline on a audio file of 1 hr duration and getting each segment of almost 30 sec each.
When I don't use BatchedInferencePipeline, segment duration is 1-2 sec and it's be…
-
after try step from readme
```
curl -X POST http://127.0.0.1:8080/v1/create/rag -F "file=@paris.txt"
```
It took 590824.84 ms = nearly 1 minute for only chunking 306 lines (91KB) file on m3 max.
…
-
One hard thing with deduplication over WORM merkle-tree datastructures like unixfs is that we are extremely limited when finding backrefs. Usually deduplication code is able to search for backrefs fre…
-
### Checked other resources
- [X] I added a very descriptive title to this issue.
- [X] I searched the LangChain.js documentation with the integrated search.
- [X] I used the GitHub search to find a …
-
1. We should check vis before serializing chunks, so we can use the right iteration (use iterate rows / use `rows_with_holes` / consider serializing columns).
2. Value chunk should be treated the sam…
-
When I run the two-line program below, I get the error. Does anyone know how to solve this? Is there a specific list of versions of the dependencies I should be using?
Thank you!!
```
Python …
-
Big games sad
-
### 🚀 The feature, motivation and pitch
Speculative decoding can achieve 50%+ latency reduction, but in vLLM it can suffer from the throughput-optimized default scheduling strategy where prefills are…
-
https://weaviate.io/blog/late-chunking