Lambda

aid: lambda-labs
name: Lambda
description: Lambda (formerly Lambda Labs) is a GPU cloud provider offering on-demand NVIDIA GPU instances,
  1-Click Clusters of 16-2,000+ interconnected H100 and B200 GPUs, long-term reserved Hyperplane capacity,
  filesystems, firewalls, and SSH key management. Lambda Cloud is accessed via a REST control-plane API,
  the web console, and SDKs/CLI tooling.
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
url: https://raw.githubusercontent.com/api-evangelist/lambda-labs/refs/heads/main/apis.yml
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.19'
type: Index
access: 3rd-Party
position: Producer
tags:
- AI
- Cloud
- Clusters
- Compute
- GPU
- Inference
- Machine Learning
apis:
- aid: lambda-labs:cloud-api
  name: Lambda Cloud API
  description: The Lambda Cloud API is the REST control plane for launching, listing, starting, stopping,
    and terminating GPU instances, managing SSH keys, firewalls, filesystems, images, and instance types.
    It supports on-demand single instances and underpins programmatic access to Lambda's GPU fleet.
  humanURL: https://docs.lambda.ai/api/cloud
  baseURL: https://cloud.lambda.ai/api/v1
  tags:
  - Cloud
  - Compute
  - Filesystems
  - Firewalls
  - GPU
  - Instances
  - REST
  - SSH Keys
  properties:
  - type: Documentation
    url: https://docs.lambda.ai/api/cloud
  - type: Documentation
    url: https://docs.lambda.ai/public-cloud/cloud-api/
- aid: lambda-labs:1-click-clusters
  name: Lambda 1-Click Clusters
  description: Lambda 1-Click Clusters provision interconnected clusters of 16 to 2,000+ NVIDIA H100 SXM
    or B200 GPUs for short-duration distributed training workloads. The product is exposed through the Lambda
    Cloud console and is integrated with the Cloud API surface.
  humanURL: https://docs.lambda.ai/public-cloud/1-click-clusters/
  tags:
  - B200
  - Clusters
  - GPU
  - H100
  - Training
  properties:
  - type: Documentation
    url: https://docs.lambda.ai/public-cloud/1-click-clusters/
- aid: lambda-labs:filesystems
  name: Lambda Cloud Filesystems
  description: Lambda Cloud Filesystems provide persistent, sharable storage attached to on-demand instances
    for datasets and checkpoints. Filesystems are managed through the Cloud API and console.
  humanURL: https://docs.lambda.ai/public-cloud/filesystems/
  tags:
  - Filesystems
  - Persistent Storage
  - Storage
  properties:
  - type: Documentation
    url: https://docs.lambda.ai/public-cloud/filesystems/
common:
- type: Website
  url: https://lambda.ai
- type: Developer
  url: https://docs.lambda.ai
- type: Documentation
  url: https://docs.lambda.ai
- type: Portal
  url: https://cloud.lambda.ai
- type: SignUp
  url: https://cloud.lambda.ai/sign-up
- type: Login
  url: https://cloud.lambda.ai/login
- type: Pricing
  url: https://lambda.ai/service/gpu-cloud
- type: Blog
  url: https://lambda.ai/blog
- type: TermsOfService
  url: https://lambda.ai/legal/terms-of-service
- type: PrivacyPolicy
  url: https://lambda.ai/legal/privacy-policy
- type: Support
  url: https://lambda.ai/contact
- type: GitHubOrganization
  url: https://github.com/LambdaLabsML
- type: LinkedIn
  url: https://www.linkedin.com/company/lambda-labs
- name: Lambda Stack
  url: https://lambda.ai/lambda-stack-deep-learning-software
  type: SDK
- type: Features
  data:
  - name: On-Demand GPU Instances
    description: Self-serve, first-come access to 1x, 2x, 4x, and 8x NVIDIA GPU virtual machines billed
      per-hour with no egress fees.
  - name: 1-Click Clusters
    description: Pre-configured clusters of 16-2,000+ interconnected H100 SXM or B200 GPUs for distributed
      training.
  - name: Hyperplane Reserved Capacity
    description: Long-term reserved GPU clusters sold via direct sales engagement for sustained training
      workloads.
  - name: Lambda Stack
    description: Pre-installed CUDA, cuDNN, PyTorch, and TensorFlow software stack across all Lambda Cloud
      instances.
  - name: Filesystems
    description: Persistent, sharable storage attached to on-demand instances for datasets and checkpoints.
- type: GPUs
  data:
  - name: NVIDIA B200 SXM6
    description: Blackwell-generation flagship GPU, listed at $6.69-$6.99/GPU/hr.
  - name: NVIDIA H100 SXM
    description: 80GB Hopper GPU on SXM, listed at $3.99-$4.29/GPU/hr.
  - name: NVIDIA GH200
    description: Grace Hopper Superchip, listed at $2.29/GPU/hr.
  - name: NVIDIA A100 SXM (80GB)
    description: 80GB Ampere GPU on SXM, listed at $2.79/GPU/hr.
  - name: NVIDIA A100 PCIe (40GB)
    description: 40GB Ampere GPU on PCIe, listed at $1.99/GPU/hr.
  - name: NVIDIA RTX A6000
    description: 48GB Ada/Ampere workstation GPU, listed at $1.09/GPU/hr.
  - name: NVIDIA Tesla V100
    description: 16GB Volta GPU, listed at $0.79/GPU/hr.
- type: Integrations
  data:
  - name: Hugging Face
    description: Reference workflows and templates for serving and training Hugging Face models on Lambda
      Cloud.
  - name: PyTorch
    description: PyTorch ships pre-installed via Lambda Stack on all instances.
  - name: TensorFlow
    description: TensorFlow ships pre-installed via Lambda Stack on all instances.
maintainers:
- FN: Kin Lane
  email: [email protected]
  url: https://apievangelist.com