feat: automatic batching

[giladgd]

Description of change

• feat: evaluate multiple sequences in parallel with automatic batching
• feat: improve automatic chat wrapper resolution
• feat: smart context shifting
• feat: improve TS types
• refactor: improve API
• build: support beta releases
• build: improve dev configurations

BREAKING CHANGE: completely new API (docs will be updated before a stable version is released)

Closes #85
Fixes #102
Fixes #94
Fixes #93
Fixes #76

Things left to do (in other PRs)

• Update documentation
• Use the smart context shifting support in LlamaChatSession
• Add contexts manager to automatically create more contexts as needed
• Improve grammar support
• Try to disable llama.cpp logs by default
• Add migration guide from v2 to v3
• Add more tests

Pull-Request Checklist

• Code is up-to-date with the master branch
• npm run format to apply eslint formatting
• npm run test passes with this change
• This pull request links relevant issues as Fixes #0000
• There are new or updated unit tests validating the change
• Documentation has been updated to reflect this change
• The new commits and pull request title follow conventions explained in pull request guidelines (PRs that do not follow this convention will not be merged)

[ido-pluto]

LGTM

[ido-pluto]

LGTM

[github-actions]

🎉 This PR is included in version 3.0.0-beta.1 🎉

The release is available on:

• npm package (@beta dist-tag)
• GitHub release
• v3.0.0-beta.1

Your semantic-release bot 📦🚀

[Madd0g]

is there a code snippet that shows how to correctly use batching?
I'm doing repetitive things in a loop and wondering how I might take advantage of this?

[giladgd]

@Madd0g There will be a better example in the documentation when version 3 leaves the beta status soon, but for now, here's a simple example:

import {fileURLToPath} from "url";
import path from "path";
import {getLlama, LlamaChatSession} from "node-llama-cpp";

const __dirname = path.dirname(fileURLToPath(import.meta.url));

const llama = await getLlama();
const model = await llama.loadModel({
    modelPath: path.join(__dirname, "models", "dolphin-2.1-mistral-7b.Q4_K_M.gguf")
});
const context = await model.createContext({
    sequences: 2
});

const sequence1 = context.getSequence();
const sequence2 = context.getSequence();

const session1 = new LlamaChatSession({
    contextSequence: sequence1
});
const session2 = new LlamaChatSession({
    contextSequence: sequence2
});

const q1 = "Hi there, how are you?";
const q2 = "How much is 6+6?";

const [
    a1,
    a2
] = await Promise.all([
    session1.prompt(q1),
    session2.prompt(q2)
]);

console.log("User: " + q1);
console.log("AI: " + a1);

console.log("User: " + q2);
console.log("AI: " + a2);

The batching is done automatically across sequences of the same context

[github-actions]

🎉 This PR is included in version 3.0.0 🎉

The release is available on:

• npm package (@latest dist-tag)
• GitHub release
• v3.0.0

Your semantic-release bot 📦🚀