llama32-3b-ultrafeedback-grpo-lr1e6-armorm
minhak-song's llama32-3b-ultrafeedback-grpo-lr1e6-armorm packs 3 billion parameters into a chat model. llama32-3b-ultrafeedback-grpo-lr1e6-armorm is an open-weights chat model with roughly 3 billion parameters.
by minhak-song · 3B parameters
Best for
- low-latency chat and routing
- request routing and triage
- text classification
Ways to use llama32-3b-ultrafeedback-grpo-lr1e6-armorm in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your minhak-song API key. osFoundry discovers llama32-3b-ultrafeedback-grpo-lr1e6-armorm 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
llama32-3b-ultrafeedback-grpo-lr1e6-armorm 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 llama32-3b-ultrafeedback-grpo-lr1e6-armorm
llama32-3b-ultrafeedback-grpo-lr1e6-armorm 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).
llama32-3b-ultrafeedback-grpo-lr1e6-armorm vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about llama32-3b-ultrafeedback-grpo-lr1e6-armorm
Is llama32-3b-ultrafeedback-grpo-lr1e6-armorm free to use?
llama32-3b-ultrafeedback-grpo-lr1e6-armorm 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 llama32-3b-ultrafeedback-grpo-lr1e6-armorm 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 llama32-3b-ultrafeedback-grpo-lr1e6-armorm need?
Approximately 2 GB at Q4 quantisation, or 8 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run llama32-3b-ultrafeedback-grpo-lr1e6-armorm locally?
Yes. llama32-3b-ultrafeedback-grpo-lr1e6-armorm is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is llama32-3b-ultrafeedback-grpo-lr1e6-armorm best at?
llama32-3b-ultrafeedback-grpo-lr1e6-armorm is well-suited to low-latency chat and routing, request routing and triage, text classification.
How do I use llama32-3b-ultrafeedback-grpo-lr1e6-armorm in osFoundry?
Paste your minhak-song API key in the key dialog (or deploy the open weights for self-hostable models), assign llama32-3b-ultrafeedback-grpo-lr1e6-armorm to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by minhak-song on April 24, 2026. Source: https://huggingface.co/minhak-song/llama32-3b-ultrafeedback-grpo-lr1e6-armorm