gpt-oss-20b-sft-trim-120b-16tasks-best-20260420
Released by jiosephlee in 2026, gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 is a 20 billion parameter chat model. gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 is an open-weights chat model with roughly 20 billion parameters.
by jiosephlee · 20B parameters
Best for
Ways to use gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your jiosephlee API key. osFoundry discovers gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 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
gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 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 gpt-oss-20b-sft-trim-120b-16tasks-best-20260420
gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 runs on a single 16GB consumer GPU (~12 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~48 GB).
gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about gpt-oss-20b-sft-trim-120b-16tasks-best-20260420
Is gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 free to use?
gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 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 gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 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 gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 need?
Approximately 12 GB at Q4 quantisation, or 48 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 locally?
Yes. gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 best at?
gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 is well-suited to text generation.
How do I use gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 in osFoundry?
Paste your jiosephlee API key in the key dialog (or deploy the open weights for self-hostable models), assign gpt-oss-20b-sft-trim-120b-16tasks-best-20260420 to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by jiosephlee on April 20, 2026. Source: https://huggingface.co/jiosephlee/gpt-oss-20b-sft-trim-120b-16tasks-best-20260420