Skip to content

feat: add hugging face loading and interactive chat #9

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
zatevakhin merged 1 commit into from
Aug 21, 2026
Merged

feat: add hugging face loading and interactive chat #9

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,9 @@ All notable changes to this project will be documented in this file.

- Checked-in raw Rust declarations for the 19-symbol stable vllm.cpp C API at ABI version 10, with header, symbol, layout, and runtime conformance checks.
- A safe API for model loading, blocking completion and streaming, raw-JSON and optional serde chat, structured output, owned sampling parameters, and concurrent request submission, cancellation, waiting, and diagnostics.
- An always-available synchronous `hf-hub` resolver for standalone GGUF files and runtime-complete sparse Safetensors snapshots, defaulting to the Hub's mutable `main` revision with an explicit branch/tag/commit override, cache/token/progress/offline controls, and no async runtime.
- Consistent local, Hugging Face GGUF, and Hugging Face Safetensors model-source arguments across every runnable example, with cache reuse and optional revisions; plus a weather extraction example and model-backed test using JSON-Schema structured output.
- A Clap-based interactive `chat` example with prompt/file startup input, retained system/user/assistant history, supported sampling controls, default streaming or blocking output, and shared local/Hugging Face resolution.
- RAII ownership for native engines, requests, completions, and strings, including callback panic containment and callback-thread-safe deferred request cleanup.
- Linux x86_64 CPU builds for bundled and system libraries with static or dynamic linking.
- Experimental bundled Linux x86_64/aarch64 build integration for CUDA, external CUTLASS, Triton AOT, and Vulkan.
Expand Down
Loading
Loading