hugo/DS4Server
1
0
Fork 0
Code Issues 7 Pull Requests Actions Packages Projects Releases Wiki Activity

feat: first cut at openai compatible server

Browse Source
This commit is contained in:
Georg Bauer
2026-07-24 21:12:48 +02:00
parent bbbe65a75f
commit d554b77b9d
18 changed files with 2612 additions and 535 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
README.md
View File

@@ -4,11 +4,15 @@ DS4Server is a native macOS coding-agent application that rewrites the
DwarfStar (`ds4`) inference engine in Rust. It uses Rust and Iced and will combine local model loading, an
OpenAI-compatible localhost endpoint, and project-scoped agent chat in one app.
DS4Server vendors and adapts the Metal kernels and Objective-C Metal glue from
[DwarfStar (`ds4`)](https://github.com/antirez/ds4). Their copyright and license
notices are retained in [`native/metal/LICENSE`](native/metal/LICENSE).
The current milestone provides a Codex-inspired project/session layout. A native
macOS folder picker selects each workspace, then the app asks for its display
name. Projects, sessions, and model preferences are persisted through Diesel in
SQLite. Open Preferences with `Command-,` to configure model, generation,
runtime, and idle-unload settings. The separate Model Manager
runtime, local endpoint, and idle-unload settings. The separate Model Manager
(`Shift-Command-M`)
lists local main and DSpark artifacts, their on-disk sizes and state, and lets
you download, resume, validate, or delete them. Rust-native background work
@@ -21,9 +25,17 @@ The selected DeepSeek V4 Flash model can run directly from a project session.
The model, KV/compressor state, 43-layer graph, sampling, and lifecycle are
owned by Rust; a fixed snapshot of the Objective-C Metal boundary and unchanged
Metal kernels is vendored and built inside this repository. Tokens stream into the chat UI, Stop cancels generation,
follow-up turns replay their role history, and the model unloads after the idle
timeout. The current graph supports 2K context while the sparse long-context
indexer and persistent transcript/KV restore remain the next engine slice.
follow-up turns reuse durable transcript and KV state, and the model unloads
after the idle timeout. The graph uses the full configured context with the
ratio-4 sparse indexer.
The app also listens on `127.0.0.1:4000` by default for `GET /v1/models` and
`POST /v1/chat/completions`; the port is configurable in Preferences. The
endpoint and local chat share the single model owner. External conversations
are client-managed and never enter the project, session, message, or transcript
database; only opaque content-addressed KV cache files are retained. Model
discovery advertises only supported main artifacts that are fully downloaded
and verified on disk.
```sh
cargo install cargo-packager --locked --version 0.11.8
@@ -35,6 +47,7 @@ State is stored at:
```text
~/Library/Application Support/DS4Server.rfc1437.de/data.sqlite3
~/Library/Application Support/DS4Server.rfc1437.de/kv-cache/
```
Deleting a project or session removes only DS4Server metadata. It never deletes