Meta-Llama-3.1-405B-Instruct-GPTQ-INT4
Built by hugging-quants, Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 is a 405 billion parameter chat model. Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 is an open-weights chat model with roughly 405 billion parameters.
by hugging-quants · 405B parameters
Best for
Ways to use Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your hugging-quants API key. osFoundry discovers Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 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
Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 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 Meta-Llama-3.1-405B-Instruct-GPTQ-INT4
Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 runs on a multi-GPU setup or H200 141GB at Q4 (~243 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~972 GB).
Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about Meta-Llama-3.1-405B-Instruct-GPTQ-INT4
Is Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 free to use?
Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 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 Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 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 Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 need?
Approximately 243 GB at Q4 quantisation, or 972 GB at full FP16 precision. Requires multi-GPU at higher quantisation.
Can I run Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 locally?
Yes. Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 best at?
Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 is well-suited to text generation.
How do I use Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 in osFoundry?
Paste your hugging-quants API key in the key dialog (or deploy the open weights for self-hostable models), assign Meta-Llama-3.1-405B-Instruct-GPTQ-INT4 to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by hugging-quants on July 20, 2024. Source: https://huggingface.co/hugging-quants/Meta-Llama-3.1-405B-Instruct-GPTQ-INT4