Kushandisa AI mamodheru kubva kuNVIDIA Tesla P40 (24 GB VRAM, Pascal architecture): LLMs, mamodheru ezvifananidzo uye zvimwe zvinofungidzirwa. GPU maserveri haasi kupihwa; taura nesu kuti uzive kana zvave kupihwa.
$ pip install vllm && vllm serve meta-llama/Llama-3-8B-Instruct
# Mufananidzo wekuisa command
AI inference zvinoreva kufambisa mamodheru akadzidziswa kuti aite zviratidzo. A GPU inference server inochengeta mufananidzo wakaiswa paGPU uye anobvunza mibvunzo nekukurumidza kupfuura CPU.
GPU inoendeswa kune yako server chete, saka latency haisi kukonzerwa neGPU mabasa evamwe vatengi.
Usashandisa Ollama, llama.cpp, Triton kana yako API. Zvimwe zvirongwa, senge vLLM, zvinoda maGPUs atsva kupfuura Pascal P40.
Kushanda masevhisi akawanda pamberi pekutakura kwako kwekutakura kana imwe GPU isina kunyatsokwana.
Kushanda imwe pakati-size model kana akawanda madiki.
AI Inference inogona kushandisa NVIDIA GPU kuti iite basa rayo. VPS.org GPU servers hazvina kuwanikwa kuti utore; taura nesu kuti uzive kana vachiri. AI Inference inowanzo GPU-yakakwira basa. Ita shuwa kuti software yako ine CUDA rutsigiro uye kuti yako driver / runtime versions inosangana nezvinodiwa zvebasa re AI Inference.
GPU servers hadzina kuwanikwa kuti utore. Nhasi unogona kuisa AI Inference pane yakajairika CPU package, uko inoita basa rakakura kupfuura pane GPU: SSH in uye shandisa pip install vllm && vllm serve meta-llama/Llama-3-8B-Instruct.
VRAM zvinodiwa AI Inference kuenderana nemhando uye batch size; tarisa chirongwa ruzivo rwezvinoreva.
GPU mutengo achaburitswa kana GPU servers anogona kupihwa. Tibate kana iwe uchida kuziva kana ichiitika.
Yega. Every VPS.org server inouya nefull root access, saka unogona kutamba zvimwe zvinhu pamwe AI Inference mukati server's RAM uye kuchengetera.
Every VPS.org server inouya nepfupi root SSH kuwanikwa, saka unogona kuisa madhiraivha, mabhuku uye AI Inference asi iwe unofanirwa.
Hapana server inouya neCUDA runtime kana NVIDIA mutyairi akaiswa, uye GPU servers hazvina kuwanikwa kuti aite kusvika parizvino. Unogona kuisa zvinyorwa AI Inference zvinoda iwe pachako.
VPS.org servers ndezvemapurojekiti akaomarara, kwete ephemeral instances: AI Inference models, configs uye data inogara paSSD storage yeserver kusvika wabvisa.
Tenga data rekushandisa pane SSD ye server yekuwedzera kuwanikwa kwenguva pfupi panguva ye AI Inference runs, uye kopa zvakapera zviwanikwa (zviyero, zvizvarwa, embeddings) kune yekuchengetera kunze kwe VPS.org; snapshots uye backups vanogara pamwechete kuchengetera cluster.
Unogona kusimudzira chero VPS.org server kune yakakura chirongwa kubva kudzora panel; resize rinotangazve uye yako AI Inference kuiswa inotora pamusoro. GPU servers hazvina kuwanikwa kusvika parizvino.
Yeah. Backups ndivo akabhadhara add-on (zuva nezuva + 30% yemari chirongwa, akachengetwa 7 mazuva; zuva nezuva + 20%, akachengetwa 28 mazuva) uye snapshots vakasununguka, izvo zvinokosha pamberi AI Inference mazuva. Both stay on the same storage cluster as the server.
Contact support within 30 days of your first purchase to request a refund. GPU servers are not sold yet, so no GPU order is involved.
GPU kugona kuri kuwanikwa chete. Tiudzai nezvebasa rako uye tichakuudza kuti ripi rinowanikwa uye ripi risiri.