v0.5.0-rc.4

v0.5.0-rc.4

v0.5.0

Release 0.5.0

[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]

Generated by cargo-workspaces

v0.5.0-rc.3

v0.5.0-rc.3

v0.5.0-rc.2

v0.5.0-rc.2

v0.5.0-rc.0

⚠️ Notice

• llama.cpp backend (CPU, Metal) now requires a redownload of gguf model due to upstream format changes: #645 ggml-org/llama.cpp#3252
• Due to indexing format changes, the ~/.tabby/index needs to be manually removed before any further runs of tabby scheduler.
• TABBY_REGISTRY is replaced with TABBY_DOWNLOAD_HOST for the github based registry implementation.

🚀 Features

• Improved dashboard UI.

🧰 Fixes and Improvements

• Cpu backend is switched to llama.cpp: #638
• add server.completion_timeout to control the code completion interface timeout: #637
• Cuda backend is switched to llama.cpp: #656
• Tokenizer implementation is switched to llama.cpp, so tabby no longer need to download additional tokenizer file: #683

💫 New Contributors

• @CrCs2O4 made their first contribution in #597
• @yusiwen made their first contribution in #620
• @gjedeer made their first contribution in #635
• @XpycT made their first contribution in #634
• @HKABIG made their first contribution in #662

Full Changelog: v0.4.0...v0.5.0

v0.4.0

🚀 Features

• Added support for golang: #553
• Added support for ruby: #597
• Supports using local directory for Repository.git_url: use file:///path/to/repo to specify a local directory.
• Introduced a new UI design for webserver.

🧰 Fixes and Improvements

• Improve snippets retrieval by dedup candidates to existing content + snippets: #582

💫 New Contributors

• @sensinsane made their first contribution in #548
• @CrCs2O4 made their first contribution in #597
• @yusiwen made their first contribution in #620

Full Changelog: v0.3.1...v0.4.0

v0.4.0-rc.0

v0.4.0-rc.0

v0.3.1

🧰 Fixes and Improvements

• Fix GPU OOM issue caused the parallelism: #541, #587
• Fix git safe directory check in docker: #569

Full Changelog: v0.3.0...v0.3.1

v0.3.1-rc.0

v0.3.1-rc.0

v0.3.0

🚀 Features

Retrieval-Augmented Code Completion Enabled by Default

The currently supported languages are:

• Rust
• Python
• JavaScript / JSX
• TypeScript / TSX

A blog series detailing the technical aspects of Retrieval-Augmented Code Completion will be published soon. Stay tuned!

Example prompt for retrieval-augmented code completion

// Path: crates/tabby/src/serve/engine.rs
// fn create_llama_engine(model_dir: &ModelDir) -> Box<dyn TextGeneration> {
//     let options = llama_cpp_bindings::LlamaEngineOptionsBuilder::default()
//         .model_path(model_dir.ggml_q8_0_file())
//         .tokenizer_path(model_dir.tokenizer_file())
//         .build()
//         .unwrap();
//
//     Box::new(llama_cpp_bindings::LlamaEngine::create(options))
// }
//
// Path: crates/tabby/src/serve/engine.rs
// create_local_engine(args, &model_dir, &metadata)
//
// Path: crates/tabby/src/serve/health.rs
// args.device.to_string()
//
// Path: crates/tabby/src/serve/mod.rs
// download_model(&args.model, &args.device)
    } else {
        create_llama_engine(model_dir)
    }
}

fn create_ctranslate2_engine(
    args: &crate::serve::ServeArgs,
    model_dir: &ModelDir,
    metadata: &Metadata,
) -> Box<dyn TextGeneration> {
    let device = format!("{}", args.device);
    let options = CTranslate2EngineOptionsBuilder::default()
        .model_path(model_dir.ctranslate2_dir())
        .tokenizer_path(model_dir.tokenizer_file())
        .device(device)
        .model_type(metadata.auto_model.clone())
        .device_indices(args.device_indices.clone())
        .build()
        .

Following the instruction in https://tabby.tabbyml.com/docs/configuration to enable repository context for your Tabby instance.

🧰 Fixes and Improvements

• Fix Issue #511 by marking ggml models as optional.
• Improve stop words handling by combining RegexSet into Regex for efficiency.

💫 New Contributors

• @sensinsane made their first contribution in #548

Full Changelog: v0.2.2...v0.3.0