gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT
Thireus's gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT packs 31 billion parameters into a chat model. gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT is an open-weights chat model with roughly 31 billion parameters.
by Thireus · 31B parameters
Best for
- low-latency chat and routing
- request routing and triage
- text classification
Ways to use gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your Thireus API key. osFoundry discovers gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT 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-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT 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-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT
gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT runs on a 24GB consumer or workstation GPU (~19 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~75 GB).
gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT
Is gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT free to use?
gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT 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-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT 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-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT need?
Approximately 19 GB at Q4 quantisation, or 75 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT locally?
Yes. gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT best at?
gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT is well-suited to low-latency chat and routing, request routing and triage, text classification.
How do I use gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT in osFoundry?
Paste your Thireus API key in the key dialog (or deploy the open weights for self-hostable models), assign gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by Thireus on April 20, 2026. Source: https://huggingface.co/Thireus/gemma-4-31B-it-THIREUS-IQ4_KS_R4-SPECIAL_SPLIT