EvoQwen2.5-VL-Retriever-3B-v1
EvoQwen2.5-VL-Retriever-3B-v1 is a 3 billion parameter chat model from ApsaraStackMaaS, released November 4, 2025. EvoQwen2.5-VL-Retriever-3B-v1 is an open-weights chat model with roughly 3 billion parameters.
by ApsaraStackMaaS · 3B parameters
Best for
- visual document retrieval
Ways to use EvoQwen2.5-VL-Retriever-3B-v1 in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your ApsaraStackMaaS API key. osFoundry discovers EvoQwen2.5-VL-Retriever-3B-v1 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
EvoQwen2.5-VL-Retriever-3B-v1 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 EvoQwen2.5-VL-Retriever-3B-v1
EvoQwen2.5-VL-Retriever-3B-v1 runs on a single 16GB consumer GPU (~2 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~8 GB).
EvoQwen2.5-VL-Retriever-3B-v1 vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about EvoQwen2.5-VL-Retriever-3B-v1
Is EvoQwen2.5-VL-Retriever-3B-v1 free to use?
EvoQwen2.5-VL-Retriever-3B-v1 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 EvoQwen2.5-VL-Retriever-3B-v1 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 EvoQwen2.5-VL-Retriever-3B-v1 need?
Approximately 2 GB at Q4 quantisation, or 8 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run EvoQwen2.5-VL-Retriever-3B-v1 locally?
Yes. EvoQwen2.5-VL-Retriever-3B-v1 is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is EvoQwen2.5-VL-Retriever-3B-v1 best at?
EvoQwen2.5-VL-Retriever-3B-v1 is well-suited to visual document retrieval.
How do I use EvoQwen2.5-VL-Retriever-3B-v1 in osFoundry?
Paste your ApsaraStackMaaS API key in the key dialog (or deploy the open weights for self-hostable models), assign EvoQwen2.5-VL-Retriever-3B-v1 to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by ApsaraStackMaaS on November 4, 2025. Source: https://huggingface.co/ApsaraStackMaaS/EvoQwen2.5-VL-Retriever-3B-v1