gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn
Released by raikonenamd in 2026, gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn is a 120 billion parameter chat model. gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn is an open-weights chat model with roughly 120 billion parameters.
by raikonenamd · 120B parameters
Best for
- complex multi-step reasoning
- agent orchestration with tool use
- long-document analysis and summarisation
Ways to use gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your raikonenamd API key. osFoundry discovers gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn 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-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn 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-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn
gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn runs on a single A100 80GB or H100 80GB at Q4 quantisation (~72 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~288 GB).
gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn 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-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn
Is gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn free to use?
gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn 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-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn 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-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn need?
Approximately 72 GB at Q4 quantisation, or 288 GB at full FP16 precision. Fits on a single A100/H100 80GB.
Can I run gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn locally?
Yes. gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn best at?
gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn is well-suited to complex multi-step reasoning, agent orchestration with tool use, long-document analysis and summarisation.
How do I use gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn in osFoundry?
Paste your raikonenamd API key in the key dialog (or deploy the open weights for self-hostable models), assign gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by raikonenamd on February 26, 2026. Source: https://huggingface.co/raikonenamd/gpt-oss-120b-w-mxfp4-a-fp8-kv-fp8-fp8attn