Tekoäly & koneoppiminen VPS

LLM- ja ML-päättely korkeatiheys-CPU:lla

Suositeltu paketti

Singularity

32 vCore EPYC · 64 GB DDR5 · 800 GB NVMe Gen4 · 10 Gbps

alkaen

$289/kk

Ota käyttöön nyt

Miksi tämä käyttötapaus

AI without GPU is possible — with the right CPUs

Not everyone needs an H100. Quantized LLM inference (GGUF in Q4/Q5), light fine-tuning of small models, scikit-learn pipelines, embedding vectorization: all of this runs remarkably well on CPU, provided you have many fast cores and enormous memory bandwidth. That's precisely the profile of our EPYC chips with 12 DDR5 channels: 460 GB/s theoretical per socket.

LLM inference on Singularity

A Llama 3.1 8B in Q5_K_M takes ~6 GB and runs at 15-20 tokens/second on 16 EPYC vCores with llama.cpp — plenty for a business chatbot or internal agent. A Mistral 7B in Q4 exceeds 25 tok/s. On Singularity (32 vCore, 64 GB), you can serve a quantized 13B or several 7-8B models in parallel behind Ollama or CPU vLLM.

DDR5 makes a measurable difference

LLM inference is memory-bandwidth-bound, not compute-bound. Our internal measurements: +38% tokens/second on our DDR5-4800 platform versus a DDR4-3200 platform with identical model and cores. That's the hardware reason two "8 vCore" VPS are not equal.

Real use cases among our customers

Support ticket classification (DistilBERT fine-tuning), RAG answer generation with pgvector embeddings, lead scoring, automatic summarization of legal documents. The common thread: batch or low-concurrency workloads, where a GPU would be wasted.

When to move to GPU

If you serve high-concurrency real-time (100+ simultaneous requests) or fine-tune models > 13B, you'll need a GPU. For everything else — and that's 90% of business use cases — a Singularity CPU is four times cheaper for equivalent service.

Pika-aloitus

  1. 1Deploy a Singularity, Ubuntu 24.04.
  2. 2Install Ollama or compile llama.cpp with AVX-512 optimizations.
  3. 3Download a Q5_K_M model (ollama pull llama3.1:8b).
  4. 4Expose the API behind Nginx + auth, or integrate via the SDK.
  5. 5Add pgvector if you're doing RAG.

Usein kysytyt kysymykset

He käyttävät sitä tuotannossa

Singularity

We host 200 of our clients' instances on Singularity. The 10 Gbps network delivers under sustained load.

TV

Théo Van der Berg

Netherlands · 21.9.2024

Vahvistettu ostos