llama-3.3-70b-instruct-fp16
Released by mlx-community in 2025, llama-3.3-70b-instruct-fp16 is a 70 billion parameter chat model. llama-3.3-70b-instruct-fp16 is an open-weights chat model with roughly 70 billion parameters.
by mlx-community · 70B parameters
Best for
Ways to use llama-3.3-70b-instruct-fp16 in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your mlx-community API key. osFoundry discovers llama-3.3-70b-instruct-fp16 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
llama-3.3-70b-instruct-fp16 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 llama-3.3-70b-instruct-fp16
llama-3.3-70b-instruct-fp16 runs on a single A100 80GB or H100 80GB at Q4 quantisation (~42 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~168 GB).
llama-3.3-70b-instruct-fp16 vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about llama-3.3-70b-instruct-fp16
Is llama-3.3-70b-instruct-fp16 free to use?
llama-3.3-70b-instruct-fp16 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 llama-3.3-70b-instruct-fp16 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 llama-3.3-70b-instruct-fp16 need?
Approximately 42 GB at Q4 quantisation, or 168 GB at full FP16 precision. Fits on a single A100/H100 80GB.
Can I run llama-3.3-70b-instruct-fp16 locally?
Yes. llama-3.3-70b-instruct-fp16 is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is llama-3.3-70b-instruct-fp16 best at?
llama-3.3-70b-instruct-fp16 is well-suited to text generation.
How do I use llama-3.3-70b-instruct-fp16 in osFoundry?
Paste your mlx-community API key in the key dialog (or deploy the open weights for self-hostable models), assign llama-3.3-70b-instruct-fp16 to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by mlx-community on March 22, 2025. Source: https://huggingface.co/mlx-community/llama-3.3-70b-instruct-fp16