Llasa-1B-Multilingual
Released by HKUSTAudio in 2025, Llasa-1B-Multilingual is a 1 billion parameter speech-and-audio model. Llasa-1B-Multilingual is an open-weights audio model with roughly 1 billion parameters.
by HKUSTAudio · 1B parameters
Best for
Ways to use Llasa-1B-Multilingual in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your HKUSTAudio API key. osFoundry discovers Llasa-1B-Multilingual automatically — assign it to a Maestro role (router, direct, orchestrator, or fallback) in the Pipeline tab and it is live in every chat. Your key, your provider account — no token markup.
Deploy a dedicated endpoint
Llasa-1B-Multilingual is open-weights — run it locally for free, or deploy a dedicated GPU endpoint in your workspace for reserved capacity with no rate limits.
Use it in a Room App
Room Apps declare AI features in their manifest, then call them with invokeAI:
import { invokeAI } from '@osfoundry/app-sdk'
// 'summarize' is an AI feature declared in your app manifest.
const result = await invokeAI('summarize', userText)
Call it from your own apps
Once a model is wired into your workspace you can host it as an API and reach it from your own services, scripts, or CI — outside osFoundry.
What hardware can run Llasa-1B-Multilingual
Llasa-1B-Multilingual runs on a single 16GB consumer GPU (~1 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~3 GB).
Llasa-1B-Multilingual vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about Llasa-1B-Multilingual
Is Llasa-1B-Multilingual free to use?
Llasa-1B-Multilingual is free to run locally on your own hardware. Hosted access through osFoundry is metered (input Free (local), output Free (local)). You can switch between local and hosted at any time.
Can I use Llasa-1B-Multilingual commercially?
Commercial use is allowed with conditions. Licence terms not specified — verify the upstream model card before commercial use. Check upstream documentation.
How much VRAM does Llasa-1B-Multilingual need?
Approximately 1 GB at Q4 quantisation, or 3 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run Llasa-1B-Multilingual locally?
Yes. Llasa-1B-Multilingual is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is Llasa-1B-Multilingual best at?
Llasa-1B-Multilingual is well-suited to text to speech.
How do I use Llasa-1B-Multilingual in osFoundry?
Paste your HKUSTAudio API key in the key dialog (or deploy the open weights for self-hostable models), assign Llasa-1B-Multilingual to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by HKUSTAudio on February 7, 2025. Source: https://huggingface.co/HKUSTAudio/Llasa-1B-Multilingual