Xuan Son Nguyen
Xuan Son Nguyen
Hello, thank you for your work. I made a fork of your repo here: https://github.com/ngxson/FindMy_Docker On the `main` branch, I added docker-compose and Dockerfile so that the project can be...
Resolves #221 In this PR, I moved source code of `hash-wasm` into vendor directory (c source code), also add compile script using emscripten. As a bonus, SIMD is also enabled,...
This PR add correct type definitions for GGUF `split.*`, and use them in `ggufAllShards` for sanity checks. The sanity checks does not really serves real-life purpose, since it is very...
Resolves #42 Resolves #43 `loadModel()` now also accepts `Blob` or `File` TODO: add example
The "next generation of node package manager" ==> https://jsr.io/
~~Data is now passing as `Uint8Array`. We can do better by using Streams: https://developer.mozilla.org/en-US/docs/Web/API/Streams_API/Using_readable_streams~~ We are now using `Blob` which already provides a `ReadableStream`.
With the introduction of heapfs, we can now do more low-level things. The idea is to load File Blob directly into wllama's heap without creating any intermediate buffer. This will...
Currently, we're using `CacheStorage` to store downloaded files. This approach has some limitations on certain browsers. Another approach woule be to use Origin private file system Ref: - https://developer.mozilla.org/en-US/docs/Web/API/File_System_API/Origin_private_file_system -...
A demo usage of emscripten WORKERFS: https://emscripten.org/docs/api_reference/Filesystem-API.html#workerfs Problem: it does not support mmap
Resolve #8581 This can now accepts new lora format introduced in #8332 The output merged tensor will be forced to f16, since `ggml_cast` does not support Q-type quants. It would...