Qwen 2.5 72B
Qwen 2.5 72B is a 72 billion parameter chat model from Alibaba, released September 19, 2024. Multilingual 72B open-weights model with strong code and math performance. Native 128K context.
by Alibaba · 72B parameters · 131K token context window
Best for
- general chat and Q&A
- code generation and review
- mathematical reasoning
- multilingual conversation
Ways to use Qwen 2.5 72B in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your Alibaba API key. osFoundry discovers Qwen 2.5 72B 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
Qwen 2.5 72B 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 Qwen 2.5 72B
Qwen 2.5 72B runs on a single A100 80GB or H100 80GB at Q4 quantisation (~44 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~173 GB).
Qwen 2.5 72B vs similar models
Licence
Qwen Licence — commercial use with attribution — Permits commercial use with attribution; some restrictions on misuse.
"Built with Qwen" attribution required in derivatives.
Frequently asked about Qwen 2.5 72B
Is Qwen 2.5 72B free to use?
Qwen 2.5 72B is free to run locally on your own hardware. Hosted access through osFoundry is metered (input Free (local) / $ 0.50 /1M, output Free (local) / $ 0.70 /1M). You can switch between local and hosted at any time.
Can I use Qwen 2.5 72B commercially?
Yes — commercial use is allowed. Permits commercial use with attribution; some restrictions on misuse. "Built with Qwen" attribution required in derivatives.
What is the context window of Qwen 2.5 72B?
Qwen 2.5 72B supports a 131K token context window.
How much VRAM does Qwen 2.5 72B need?
Approximately 44 GB at Q4 quantisation, or 173 GB at full FP16 precision. Fits on a single A100/H100 80GB.
Can I run Qwen 2.5 72B locally?
Yes. Qwen 2.5 72B is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is Qwen 2.5 72B best at?
Qwen 2.5 72B is well-suited to general chat and Q&A, code generation and review, mathematical reasoning.
How do I use Qwen 2.5 72B in osFoundry?
Paste your Alibaba API key in the key dialog (or deploy the open weights for self-hostable models), assign Qwen 2.5 72B to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by Alibaba on September 19, 2024.