Configuration matérielle requise pour TaskingAI : RAM, processeur, stockage et pile d’IA
Planifiez le matériel TaskingAI pour la RAM, le CPU, les services Docker Compose, les données RAG, les API de modèles et les recommandations pratiques de déploiement sur ZimaOS.
TaskingAI hardware requirements at a glance
Size TaskingAI from verified upstream requirements first, then add headroom for workload and persistent data.
- RAM
- No universal official minimum published
- CPU
- No universal official minimum published
- Storage
- Plan persistent databases, vector/RAG data, service state, logs, and backups
- Deployment
- Docker + Docker Compose + Git
- Client runtime
- Python 3.8+ for the SDK
- Best Zima starting point
- ZimaBoard 2 1664 for a modest API/RAG stack using external model providers
From official requirements to the right setup
Start with current upstream facts for TaskingAI, then size the actual workload rather than copying generic container or app-store labels.
-
Official requirements
-
Confirm your needs
-
Leave room to grow
-
Run it on ZimaOS
Check every playback client
- Concurrent users
- RAG collection size
- Vector/search workload
- External vs local LLM
- Multi-tenant use
- Compose service count
- Database backup
- API concurrency
Official minimum requirements
TaskingAI documents Docker Compose deployment and development/client runtimes but does not publish one universal numerical whole-host CPU, RAM, or disk minimum.
Do not write a single TaskingAI minimum based on an LLM model. The orchestration platform and local inference are separate workloads; model size, quantization, context, and backend determine local inference hardware.
| Requirement | Official minimum | What this supports |
|---|---|---|
| No universal numerical minimum published | Depends on API/RAG workload and whether inference is local. | |
| No universal numerical minimum published | Compose services, RAG data, and concurrency determine memory use. | |
| Docker and Docker Compose | Required for the documented self-hosting flow. | |
| Required for the documented Community Edition quick start | Used to clone the project. | |
| Python 3.8 or newer | Applies to the client SDK test environment. | |
| 8080 | Documented Nginx mapping for the Community Edition console. | |
| Not universally required | Only separately self-hosted model inference may require or strongly benefit from GPU. |
When to upgrade your hardware
RAG/database pressure
API concurrency
Local inference added
Plan hardware growth with confidence
SSD service data
Keep databases, vector/RAG state, and service data on SSD.
SSD/NVMeMore RAM
Useful for a larger Compose stack and RAG/database cache.
ZimaBoard 2 1664 or higherStronger CPU
Useful for higher API/RAG concurrency.
ZimaCube 2 ProSeparate AI acceleration
Use GPU-capable hardware only when local inference actually requires it.
Dedicated GPU-capable nodeCan it run on ZimaOS?
Install TaskingAI from the ZimaOS App Store
A specific public TaskingAI App Store page is verified, so this page uses the one-click route.
Open TaskingAI in the ZimaOS App Store ↗Use TaskingAI's documented Docker Compose stack
The Community Edition self-hosting guide is the source of truth for service relationships and deployment.
Open TaskingAI self-hosting docs ↗Size local LLM inference separately
TaskingAI can orchestrate model providers; a local model's RAM/VRAM requirement is not the platform's universal minimum.
Open TaskingAI project documentation ↗Zima hardware for TaskingAI
TaskingAI should be sized as an AI application platform plus its stateful services. Local model inference must be treated as a separate workload.
Choose by workload
Prioritize RAM and SSD storage for the multi-service stack.
Use stronger CPU for the platform and size local models separately.
Do not recommend GPU hardware solely because the app category is AI; GPU is justified only when local inference actually runs on this host.
| Zima hardware | Best for | Example workload | Core configuration | Recommended boundary | Next step |
|---|---|---|---|---|---|
| ZimaBoard 2 1664 | Memory-heavier services and larger multi-container stacks | Workloads that benefit from 16 GB RAM without needing a much faster CPU |
|
CPU throughput remains the same 4-core N150, so compute-heavy workloads can still be CPU-limited | Get Now |
| ZimaCube 2 Pro | CPU-heavier, database-heavier, and higher-concurrency workloads | Larger SQL/search workloads, many users, or broader server consolidation |
|
16 GB RAM can still constrain very large AI or enterprise workloads | Get Now |
What the Press Says
Highlights from trusted reviewers worldwide.
“ZimaCube 2: Not just another NAS, tested with 25TB storage, local AI agents, 4K transcoding, and real homelab workflows.”Read full review
“The ZimaBoard 2 is a compact x86 server board that can be turned into a mini NAS, home server, media box, or self-hosting hub.”Read full review
“ZimaCube 2: A modern, high-performance NAS with plenty of room to grow—built for users who want more than basic storage.”Read full review
“Coverage focused on ZimaCube 2's open hardware foundation, no monthly fee, and self-hosting flexibility.”Read full review
Loved by the Community
Stories and reviews from people who build with Zima every day.
Zima Blade Little yet Powerful
Maybe I am not digital natives but I live with PCs since 12 years old in 1984 when IBM PC clone come to my home. Many years have passed and many operating system I've tried. For me Zima blade and CasaOS was a quantum leap for home PC enthusiast and server lab machine to make me stay curious and relevant for this era.
Very good!!
I use ZimaCube Pro as 5th Proxmox cluster node. It runs several VMs and containers, including a VM with GPU passthrough to run a self-hosted LLM. A specific LXC container runs a Samba server for NAS capabilities using four of six RAID 6 SATA HDDs with ZFS.
Great innovation for mini server!
It is very useful and makes a powerful mini server for many purposes, including university and college students in engineering and electronics. Thank you so much for making this server.
Avaliação ZimaBoard 2
Construí um servidor de uso pessoal. O desempenho está muito bom e funciona perfeitamente onde quer que eu esteja. A surpresa é não dependermos de grandes estruturas para termos nosso próprio servidor de dados. Como iniciante, estou gostando bastante do ZimaOS, pois ele é simples e eficiente.
Frequently asked questions
How much RAM does TaskingAI need?
Upstream does not publish one universal whole-host RAM minimum.
How many CPU cores does TaskingAI need?
No universal numerical CPU minimum is published.
How is TaskingAI self-hosted?
The official Community Edition guide uses Docker Compose.
Which Python version is required?
The documented client SDK test environment requires Python 3.8 or newer.
Does TaskingAI require a GPU?
Not universally. A GPU may be needed or helpful only for a separately self-hosted local model.
Can ZimaBoard 2 1664 run TaskingAI?
It is a practical starting point for a modest stack using external model providers and moderate RAG/API workloads.
What changes when I run a local LLM?
RAM, VRAM, CPU/GPU backend, model size, quantization, and context length become separate sizing variables.
When should I choose ZimaCube 2 Pro?
For higher service concurrency, larger RAG/database workloads, or broader server consolidation.
