InternVL2-Llama3-76B
Released by OpenGVLab in 2024, InternVL2-Llama3-76B is a 76 billion parameter image-generation model. InternVL2-Llama3-76B is an open-weights image model with roughly 76 billion parameters.
by OpenGVLab · 76B parameters
Best for
Ways to use InternVL2-Llama3-76B in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your OpenGVLab API key. osFoundry discovers InternVL2-Llama3-76B 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
InternVL2-Llama3-76B 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 InternVL2-Llama3-76B
InternVL2-Llama3-76B runs on a single A100 80GB or H100 80GB at Q4 quantisation (~46 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~183 GB).
InternVL2-Llama3-76B vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about InternVL2-Llama3-76B
Is InternVL2-Llama3-76B free to use?
InternVL2-Llama3-76B 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 InternVL2-Llama3-76B 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 InternVL2-Llama3-76B need?
Approximately 46 GB at Q4 quantisation, or 183 GB at full FP16 precision. Fits on a single A100/H100 80GB.
Can I run InternVL2-Llama3-76B locally?
Yes. InternVL2-Llama3-76B is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is InternVL2-Llama3-76B best at?
InternVL2-Llama3-76B is well-suited to image text to text.
How do I use InternVL2-Llama3-76B in osFoundry?
Paste your OpenGVLab API key in the key dialog (or deploy the open weights for self-hostable models), assign InternVL2-Llama3-76B to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by OpenGVLab on July 15, 2024. Source: https://huggingface.co/OpenGVLab/InternVL2-Llama3-76B