GigaChat3.1-702B-A36B
Released by ai-sage in 2026, GigaChat3.1-702B-A36B is a 702 billion parameter chat model. GigaChat3.1-702B-A36B is an open-weights chat model with roughly 702 billion parameters.
by ai-sage · 702B parameters
Best for
Ways to use GigaChat3.1-702B-A36B in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your ai-sage API key. osFoundry discovers GigaChat3.1-702B-A36B 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
GigaChat3.1-702B-A36B 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 GigaChat3.1-702B-A36B
GigaChat3.1-702B-A36B runs on a multi-GPU setup or H200 141GB at Q4 (~422 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~1685 GB).
GigaChat3.1-702B-A36B vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about GigaChat3.1-702B-A36B
Is GigaChat3.1-702B-A36B free to use?
GigaChat3.1-702B-A36B 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 GigaChat3.1-702B-A36B 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 GigaChat3.1-702B-A36B need?
Approximately 422 GB at Q4 quantisation, or 1685 GB at full FP16 precision. Requires multi-GPU at higher quantisation.
Can I run GigaChat3.1-702B-A36B locally?
Yes. GigaChat3.1-702B-A36B is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is GigaChat3.1-702B-A36B best at?
GigaChat3.1-702B-A36B is well-suited to text generation.
How do I use GigaChat3.1-702B-A36B in osFoundry?
Paste your ai-sage API key in the key dialog (or deploy the open weights for self-hostable models), assign GigaChat3.1-702B-A36B to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by ai-sage on March 21, 2026. Source: https://huggingface.co/ai-sage/GigaChat3.1-702B-A36B