메뉴
Search for available GPU machines on Vast.ai. Use when looking for GPUs to rent, comparing GPU pricing, or finding machines that match specific requirements.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Manage Vast.ai autoscaling endpoints and worker groups for production deployments. Use when setting up auto-scaling GPU inference, managing worker pools, or deploying services.
Launch a GPU instance on Vast.ai. Use when the user wants to create, start, or spin up a new GPU machine for training, inference, or development.
Manage Vast.ai GPU instances — show status, start, stop, destroy, SSH, execute commands, view logs, copy files, take snapshots. Use when the user wants to check on, connect to, transfer files, or control their GPU instances.
Manage Vast.ai persistent storage volumes — search, create, delete, clone, and attach volumes to instances. Use for persistent data across instance lifecycles.
Run a job on a Vast.ai GPU instance end-to-end: find a GPU, launch it, execute the job, monitor it, and destroy the instance when done. Use when the user wants to run a training job, inference task, script, or any workload on a remote GPU.
Set up Vast.ai CLI — configure API key, verify installation, set up SSH keys, and check account. Use when the user needs to get started with Vast.ai or is having authentication issues.
| name | search-gpus |
| description | Search for available GPU machines on Vast.ai. Use when looking for GPUs to rent, comparing GPU pricing, or finding machines that match specific requirements. |
| argument-hint | [gpu-type or requirements] |
| allowed-tools | Bash |
Help the user find the best GPU offers on Vast.ai.
$ARGUMENTS
Understand requirements: Key dimensions to determine:
Build the search query from the mapping below.
Run the search and present results clearly.
Help interpret: Highlight best options based on user priorities.
| User says | Query field |
|---|---|
| "4090", "RTX 4090" | gpu_name=RTX_4090 |
| "A100 80GB" | gpu_name=A100 gpu_ram>=80 |
| "H100" | gpu_name=H100 |
| "4 GPUs", "multi-GPU" | num_gpus>=4 |
| "at least 48GB VRAM" | gpu_total_ram>=48 |
| "under $1/hr", "cheap" | dph_total<1.0 |
| "reliable" | reliability>0.95 |
| "US only" | geolocation=US |
| "datacenter" | datacenter=true |
| "spot", "interruptible" | add -b flag |
| "reserved" | add -r flag |
| "NVLink" | bw_nvlink>0 |
| "fast internet" | inet_down>500 |
Always include reliability>0.9 unless the user wants cheaper unreliable machines.
Default sort: -o 'dph_total' (cheapest first).
# Cheap single 4090
vastai search offers 'gpu_name=RTX_4090 num_gpus=1 reliability>0.9' -o 'dph_total'
# 8x H100 for large training
vastai search offers 'gpu_name=H100 num_gpus=8 reliability>0.95' -o 'dph_total'
# Any GPU with 80GB+ VRAM in US datacenters
vastai search offers 'gpu_total_ram>=80 geolocation=US datacenter=true reliability>0.95' -o 'dph_total'
# Budget multi-GPU fine-tuning
vastai search offers 'num_gpus>=2 gpu_ram>=24 dph_total<1.5 reliability>0.9' -o 'dph_total'
# Spot pricing for H100s
vastai search offers 'gpu_name=H100 num_gpus>=4' -b -o 'dph_total'
After running the search: