pip install vastai
vastai set api-key YOUR_API_KEY

[
  {
    "instance_id": 33799165,
    "offer_id": 25831376,
    "gpu_name": "RTX_3060",
    "num_gpus": 1,
    "dph": 0.0414,
    "ssh_url": "ssh://root@1.208.108.242:58955",
    "ssh_host": "1.208.108.242",
    "ssh_port": 58955,
    "created_at": "2026-03-29T21:12:00Z",
    "status": "running",
    "experiment": "exp01_baseline",
    "estimated_hours": 4.0,
    "estimated_cost": 0.17
  }
]

undefined

The output is a table with columns: `ID`, `CUDA`, `N` (GPU count), `Model`, `PCIE`, `cpu_ghz`, `vCPUs`, `RAM`, `Disk`, `$/hr`, `DLP` (deep learning perf), `score`, `NV Driver`, `Net_up`, `Net_down`, `R` (reliability %), `Max_Days`, `mach_id`, `status`, `host_id`, `ports`, `country`.

The **first column (`ID`)** is the offer ID needed for `vastai create instance`.

**Step 4: Present Cost-Optimized Options**

Present **3 options** to the user, ranked by estimated total cost:

**Key presentation rules:**
- Always show **estimated total cost** ($/hr × estimated hours), not just $/hr
- Faster GPUs have shorter estimated hours (scale by relative FLOPS)
- Flag if a cheap option has reliability < 0.97 ("budget pick — 3% chance of interruption")
- If task is small (<1 hour), recommend interruptible pricing for even lower cost
- If no offers meet VRAM requirements, explain why and suggest alternatives (e.g., multi-GPU, quantization)

**Relative speed scaling (approximate, for estimating hours across GPU tiers):**

| GPU | Relative Speed (FP16) |
|-----|-----------------------:|
| RTX 3060 | 0.5× |
| RTX 3090 | 1.0× |
| RTX 4090 | 1.6× |
| A5000 | 0.9× |
| A6000 | 1.1× |
| L40S | 1.5× |
| A100 SXM | 2.0× |
| H100 SXM | 3.3× |

Use these to scale the base estimated hours across offers.

vastai create instance <OFFER_ID> \
  --image <DOCKER_IMAGE> \
  --disk <DISK_GB> \
  --ssh \
  --direct \
  --onstart-cmd "apt-get update && apt-get install -y git screen rsync"

Started. {'success': True, 'new_contract': 33799165, 'instance_api_key': '...'}

vastai show instances --raw | python3 -c "
import sys, json
instances = json.load(sys.stdin)
for inst in instances:
    if inst['id'] == <INSTANCE_ID>:
        print(inst['actual_status'])
"

vastai ssh-url <INSTANCE_ID>

ssh -o StrictHostKeyChecking=no -o ConnectTimeout=15 -p <PORT> root@<HOST> "nvidia-smi && echo 'CONNECTION_OK'"

Vast.ai instance ready:
- Instance ID: <ID>
- GPU: <GPU_NAME> x <NUM_GPUS>
- Cost: $<DPH>/hr (estimated total: ~$<TOTAL>)
- SSH: ssh -p <PORT> root@<HOST>
- Docker: <IMAGE>

To deploy: /run-experiment (will auto-detect this instance)
To destroy when done: /vast-gpu destroy <ID>

ssh -p <PORT> root@<HOST> "pip install -q wandb tensorboard scipy scikit-learn pandas"

scp -P <PORT> requirements.txt root@<HOST>:/workspace/
ssh -p <PORT> root@<HOST> "pip install -q -r /workspace/requirements.txt"

rsync -avz -e "ssh -p <PORT>" \
  --include='*.py' --include='*.yaml' --include='*.yml' --include='*.json' \
  --include='*.txt' --include='*.sh' --include='*/' \
  --exclude='*.pt' --exclude='*.pth' --exclude='*.ckpt' \
  --exclude='__pycache__' --exclude='.git' --exclude='data/' \
  --exclude='wandb/' --exclude='outputs/' \
  ./ root@<HOST>:/workspace/project/

ssh -p <PORT> root@<HOST> "cd /workspace/project && python -c 'import torch; print(f\"PyTorch {torch.__version__}, CUDA: {torch.cuda.is_available()}, GPUs: {torch.cuda.device_count()}\")'"

ssh -p <PORT> root@<HOST> "ls /workspace/project/results/ 2>/dev/null || echo 'NO_RESULTS_DIR'"

rsync -avz -e "ssh -p <PORT>" root@<HOST>:/workspace/project/results/ ./results/

scp -P <PORT> root@<HOST>:/workspace/*.log ./logs/ 2>/dev/null

vastai destroy instance <INSTANCE_ID>

Instance <ID> destroyed.
- Duration: ~X.X hours
- Actual cost: ~$X.XX (estimated was $Y.YY)
- Results downloaded to: ./results/

vastai show instances

undefined

The skill analyzes experiment scripts and plans to determine what GPU to rent. No need to specify GPU model, VRAM, or instance count.

/run-experiment "train model"       ← detects gpu: vast, calls /vast-gpu provision
  ↳ /vast-gpu provision             ← analyzes task, presents options with cost
  ↳ user picks option               ← rent + setup + deploy
  ↳ /vast-gpu destroy               ← auto-destroy when done (if auto_destroy: true)

/vast-gpu provision                 ← manual: analyze task + show options
/vast-gpu rent <offer_id>           ← manual: rent a specific offer
/vast-gpu list                      ← show active instances
/vast-gpu destroy <instance_id>     ← tear down, stop billing
/vast-gpu destroy-all               ← tear down everything