TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head
Released by nm-testing in 2026, TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head is a 1 billion parameter chat model. TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head is an open-weights chat model with roughly 1 billion parameters.
by nm-testing · 1B parameters
Best for
- low-latency chat and routing
- request routing and triage
- text classification
Ways to use TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your nm-testing API key. osFoundry discovers TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head 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
TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head 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 TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head
TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head runs on a single 16GB consumer GPU (~1 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~3 GB).
TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head
Is TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head free to use?
TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head 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 TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head 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 TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head need?
Approximately 1 GB at Q4 quantisation, or 3 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head locally?
Yes. TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head best at?
TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head is well-suited to low-latency chat and routing, request routing and triage, text classification.
How do I use TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head in osFoundry?
Paste your nm-testing API key in the key dialog (or deploy the open weights for self-hostable models), assign TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by nm-testing on January 14, 2026. Source: https://huggingface.co/nm-testing/TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head