메뉴
Reserved and on-demand GPU cloud instances for ML training and inference. Use when you need dedicated GPU instances with simple SSH access, persistent filesystems, or high-performance multi-node clusters for large-scale training.
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
Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat), improving performance beyond single models, or experimenting rapidly with model variants. Covers SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, and production deployment strategies.
Compiles any research input — PDF papers, GitHub repositories, experiment logs, code directories, or raw notes — into a complete Agent-Native Research Artifact (ARA) with cognitive layer (claims, concepts, heuristics), physical layer (configs, code stubs), exploration graph, and grounded evidence. Use when ingesting a paper or codebase into a structured, machine-executable knowledge package, building an ARA from scratch, or converting research outputs into a falsifiable, agent-traversable form.
Records research provenance as a post-task epilogue, scanning conversation history at the end of a coding or research session to extract decisions, experiments, dead ends, claims, heuristics, and pivots, and writing them into the ara/ directory with user-vs-AI provenance tags. Use as a session epilogue — never during execution — to maintain a faithful, auditable trace of how a research project actually evolved.
Performs ARA Seal Level 2 semantic epistemic review on Agent-Native Research Artifacts, scoring six dimensions (evidence relevance, falsifiability, scope calibration, argument coherence, exploration integrity, methodological rigor) and producing a constructive, severity-ranked report with a Strong Accept-to-Reject recommendation. Use after Level 1 structural validation passes, when an ARA needs an objective epistemic critique before publication or release.
Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. For systems venues (OSDI, NSDI, ASPLOS, SOSP), use systems-paper-writing instead.
Generates conference presentation slides (Beamer LaTeX PDF and editable PPTX) from a compiled paper with speaker notes and talk script. Use when preparing oral talks, spotlight presentations, or invited talks for ML and systems conferences.
| name | lambda-labs-gpu-cloud |
| description | Reserved and on-demand GPU cloud instances for ML training and inference. Use when you need dedicated GPU instances with simple SSH access, persistent filesystems, or high-performance multi-node clusters for large-scale training. |
| version | 1.0.0 |
| author | Orchestra Research |
| license | MIT |
| tags | ["Infrastructure","GPU Cloud","Training","Inference","Lambda Labs"] |
| dependencies | ["lambda-cloud-client>=1.0.0"] |
Comprehensive guide to running ML workloads on Lambda Labs GPU cloud with on-demand instances and 1-Click Clusters.
Use Lambda Labs when:
Key features:
Use alternatives instead:
# Get instance IP from console
ssh ubuntu@<INSTANCE-IP>
# Or with specific key
ssh -i ~/.ssh/lambda_key ubuntu@<INSTANCE-IP>
| GPU | VRAM | Price/GPU/hr | Best For |
|---|---|---|---|
| B200 SXM6 | 180 GB | $4.99 | Largest models, fastest training |
| H100 SXM | 80 GB | $2.99-3.29 | Large model training |
| H100 PCIe | 80 GB | $2.49 | Cost-effective H100 |
| GH200 | 96 GB | $1.49 | Single-GPU large models |
| A100 80GB | 80 GB | $1.79 | Production training |
| A100 40GB | 40 GB | $1.29 | Standard training |
| A10 | 24 GB | $0.75 | Inference, fine-tuning |
| A6000 | 48 GB | $0.80 | Good VRAM/price ratio |
| V100 | 16 GB | $0.55 | Budget training |
8x GPU: Best for distributed training (DDP, FSDP)
4x GPU: Large models, multi-GPU training
2x GPU: Medium workloads
1x GPU: Fine-tuning, inference, development
All instances come with Lambda Stack pre-installed:
# Included software
- Ubuntu 22.04 LTS
- NVIDIA drivers (latest)
- CUDA 12.x
- cuDNN 8.x
- NCCL (for multi-GPU)
- PyTorch (latest)
- TensorFlow (latest)
- JAX
- JupyterLab
# Check GPU
nvidia-smi
# Check PyTorch
python -c "import torch; print(torch.cuda.is_available())"
# Check CUDA version
nvcc --version
pip install lambda-cloud-client
import os
import lambda_cloud_client
# Configure with API key
configuration = lambda_cloud_client.Configuration(
host="https://cloud.lambdalabs.com/api/v1",
access_token=os.environ["LAMBDA_API_KEY"]
)
with lambda_cloud_client.ApiClient(configuration) as api_client:
api = lambda_cloud_client.DefaultApi(api_client)
# Get available instance types
types = api.instance_types()
for name, info in types.data.items():
print(f"{name}: {info.instance_type.description}")
from lambda_cloud_client.models import LaunchInstanceRequest
request = LaunchInstanceRequest(
region_name="us-west-1",
instance_type_name="gpu_1x_h100_sxm5",
ssh_key_names=["my-ssh-key"],
file_system_names=["my-filesystem"], # Optional
name="training-job"
)
response = api.launch_instance(request)
instance_id = response.data.instance_ids[0]
print(f"Launched: {instance_id}")
instances = api.list_instances()
for instance in instances.data:
print(f"{instance.name}: {instance.ip} ({instance.status})")
from lambda_cloud_client.models import TerminateInstanceRequest
request = TerminateInstanceRequest(
instance_ids=[instance_id]
)
api.terminate_instance(request)
from lambda_cloud_client.models import AddSshKeyRequest
# Add SSH key
request = AddSshKeyRequest(
name="my-key",
public_key="ssh-rsa AAAA..."
)
api.add_ssh_key(request)
# List keys
keys = api.list_ssh_keys()
# Delete key
api.delete_ssh_key(key_id)
curl -u $LAMBDA_API_KEY: \
https://cloud.lambdalabs.com/api/v1/instance-types | jq
curl -u $LAMBDA_API_KEY: \
-X POST https://cloud.lambdalabs.com/api/v1/instance-operations/launch \
-H "Content-Type: application/json" \
-d '{
"region_name": "us-west-1",
"instance_type_name": "gpu_1x_h100_sxm5",
"ssh_key_names": ["my-key"]
}' | jq
curl -u $LAMBDA_API_KEY: \
-X POST https://cloud.lambdalabs.com/api/v1/instance-operations/terminate \
-H "Content-Type: application/json" \
-d '{"instance_ids": ["<INSTANCE-ID>"]}' | jq
Filesystems persist data across instance restarts:
# Mount location
/lambda/nfs/<FILESYSTEM_NAME>
# Example: save checkpoints
python train.py --checkpoint-dir /lambda/nfs/my-storage/checkpoints
Filesystems must be attached at instance launch time:
file_system_names in launch request# Store on filesystem (persists)
/lambda/nfs/storage/
├── datasets/
├── checkpoints/
├── models/
└── outputs/
# Local SSD (faster, ephemeral)
/home/ubuntu/
└── working/ # Temporary files
# Generate key locally
ssh-keygen -t ed25519 -f ~/.ssh/lambda_key
# Add public key to Lambda console
# Or via API
# On instance, add more keys
echo 'ssh-rsa AAAA...' >> ~/.ssh/authorized_keys
# On instance
ssh-import-id gh:username
# Forward Jupyter
ssh -L 8888:localhost:8888 ubuntu@<IP>
# Forward TensorBoard
ssh -L 6006:localhost:6006 ubuntu@<IP>
# Multiple ports
ssh -L 8888:localhost:8888 -L 6006:localhost:6006 ubuntu@<IP>
# On instance
jupyter lab --ip=0.0.0.0 --port=8888
# From local machine with tunnel
ssh -L 8888:localhost:8888 ubuntu@<IP>
# Open http://localhost:8888
# SSH to instance
ssh ubuntu@<IP>
# Clone repo
git clone https://github.com/user/project
cd project
# Install dependencies
pip install -r requirements.txt
# Train
python train.py --epochs 100 --checkpoint-dir /lambda/nfs/storage/checkpoints
# train_ddp.py
import torch
import torch.distributed as dist
from torch.nn.parallel import DistributedDataParallel as DDP
def main():
dist.init_process_group("nccl")
rank = dist.get_rank()
device = rank % torch.cuda.device_count()
model = MyModel().to(device)
model = DDP(model, device_ids=[device])
# Training loop...
if __name__ == "__main__":
main()
# Launch with torchrun (8 GPUs)
torchrun --nproc_per_node=8 train_ddp.py
import os
checkpoint_dir = "/lambda/nfs/my-storage/checkpoints"
os.makedirs(checkpoint_dir, exist_ok=True)
# Save checkpoint
torch.save({
'epoch': epoch,
'model_state_dict': model.state_dict(),
'optimizer_state_dict': optimizer.state_dict(),
'loss': loss,
}, f"{checkpoint_dir}/checkpoint_{epoch}.pt")
High-performance Slurm clusters with:
# On Slurm cluster
srun --nodes=4 --ntasks-per-node=8 --gpus-per-node=8 \
torchrun --nnodes=4 --nproc_per_node=8 \
--rdzv_backend=c10d --rdzv_endpoint=$MASTER_ADDR:29500 \
train.py
# Find private IP
ip addr show | grep 'inet '
# 1. Launch 8x H100 instance with filesystem
# 2. SSH and setup
ssh ubuntu@<IP>
pip install transformers accelerate peft
# 3. Download model to filesystem
python -c "
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained('meta-llama/Llama-2-7b-hf')
model.save_pretrained('/lambda/nfs/storage/models/llama-2-7b')
"
# 4. Fine-tune with checkpoints on filesystem
accelerate launch --num_processes 8 train.py \
--model_path /lambda/nfs/storage/models/llama-2-7b \
--output_dir /lambda/nfs/storage/outputs \
--checkpoint_dir /lambda/nfs/storage/checkpoints
# 1. Launch A10 instance (cost-effective for inference)
# 2. Run inference
python inference.py \
--model /lambda/nfs/storage/models/fine-tuned \
--input /lambda/nfs/storage/data/inputs.jsonl \
--output /lambda/nfs/storage/data/outputs.jsonl
| Task | Recommended GPU |
|---|---|
| LLM fine-tuning (7B) | A100 40GB |
| LLM fine-tuning (70B) | 8x H100 |
| Inference | A10, A6000 |
| Development | V100, A10 |
| Maximum performance | B200 |
| Issue | Solution |
|---|---|
| Instance won't launch | Check region availability, try different GPU |
| SSH connection refused | Wait for instance to initialize (3-15 min) |
| Data lost after terminate | Use persistent filesystems |
| Slow data transfer | Use filesystem in same region |
| GPU not detected | Reboot instance, check drivers |