gemma-3-12b-it-heretic-v2_fp8_e4m3fn

Built by lynaNSFW, gemma-3-12b-it-heretic-v2_fp8_e4m3fn is a 12 billion parameter image-generation model. gemma-3-12b-it-heretic-v2_fp8_e4m3fn is an open-weights image model with roughly 12 billion parameters.

by lynaNSFW · 12B parameters

Best for

text to image

Ways to use gemma-3-12b-it-heretic-v2_fp8_e4m3fn in osFoundry

Connect with your own key (BYOK)

Open the key dialog and paste your lynaNSFW API key. osFoundry discovers gemma-3-12b-it-heretic-v2_fp8_e4m3fn 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

gemma-3-12b-it-heretic-v2_fp8_e4m3fn 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 gemma-3-12b-it-heretic-v2_fp8_e4m3fn

gemma-3-12b-it-heretic-v2_fp8_e4m3fn runs on a single 16GB consumer GPU (~8 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~29 GB).

gemma-3-12b-it-heretic-v2_fp8_e4m3fn vs similar models

Model	Org	Params	Context	Input price	Self-host
gemma-3-12b-it-heretic-v2_fp8_e4m3fn	lynaNSFW	12B	—	Free (local)	Yes
pixtral-12b-240910	mistral-community	12B	—	Free (local)	Yes
mlabonne_gemma-3-12b-it-abliterated-GGUF	bartowski	12B	—	Free (local)	Yes
gemma-3-12b-it-bnb-4bit	unsloth	12B	—	Free (local)	Yes

Licence

Unspecified — Licence terms not specified — verify the upstream model card before commercial use.

Check upstream documentation.

Frequently asked about gemma-3-12b-it-heretic-v2_fp8_e4m3fn

Is gemma-3-12b-it-heretic-v2_fp8_e4m3fn free to use?

gemma-3-12b-it-heretic-v2_fp8_e4m3fn 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 gemma-3-12b-it-heretic-v2_fp8_e4m3fn 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 gemma-3-12b-it-heretic-v2_fp8_e4m3fn need?

Approximately 8 GB at Q4 quantisation, or 29 GB at full FP16 precision. Fits on a single 24GB consumer GPU.

Can I run gemma-3-12b-it-heretic-v2_fp8_e4m3fn locally?

Yes. gemma-3-12b-it-heretic-v2_fp8_e4m3fn is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.

What is gemma-3-12b-it-heretic-v2_fp8_e4m3fn best at?

gemma-3-12b-it-heretic-v2_fp8_e4m3fn is well-suited to text to image.

How do I use gemma-3-12b-it-heretic-v2_fp8_e4m3fn in osFoundry?

Paste your lynaNSFW API key in the key dialog (or deploy the open weights for self-hostable models), assign gemma-3-12b-it-heretic-v2_fp8_e4m3fn to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.

Published by lynaNSFW on April 30, 2026. Source: https://huggingface.co/lynaNSFW/gemma-3-12b-it-heretic-v2_fp8_e4m3fn