@chassis-cloud/sdk

Official TypeScript / JavaScript client for the Chassis public API (/api/v1).

Default base URL: https://chassis.okeymeta.com.ng/api/v1

npm: https://www.npmjs.com/package/@chassis-cloud/sdk

Install

npm install @chassis-cloud/sdk

Auth

Create an API key in the Chassis console under API Keys. Keys look like chs_… and are shown once. Send them as:

Authorization: Bearer chs_YOUR_KEY

Quickstart

import { Chassis } from '@chassis-cloud/sdk'

const chassis = new Chassis({
  apiKey: process.env.CHASSIS_API_KEY!,
})

const gpus = await chassis.listGpus()
console.log(gpus.map((g) => `${g.displayName} $${g.pricePerHourUsd}/hr`))

const instance = await chassis.spinUp({
  gpuSkuId: gpus[0].id,
  name: 'gpu-host-01',
  imageName: 'ghcr.io/YOUR_ORG/your-gpu-app:latest',
  ports: '8080/http,22/tcp',
})

Example: list available GPUs

const gpus = await chassis.listGpus()

for (const gpu of gpus) {
  console.log({
    id: gpu.id,
    name: gpu.displayName,
    memoryGb: gpu.memoryGb,
    pricePerHourUsd: gpu.pricePerHourUsd,
    spotPricePerHourUsd: gpu.spotPricePerHourUsd,
    stockStatus: gpu.stockStatus,
    secureAvailable: gpu.secureAvailable,
  })
}

const gpu = gpus.find((g) => g.displayName.includes('4090')) ?? gpus[0]
// use gpu.id as gpuSkuId on createInstance / spinUp / createCluster / createEndpoint

Example: host a GPU service

Any CUDA app — APIs, media tools, notebooks, batch workers — not only training.

const gpus = await chassis.listGpus()

const instance = await chassis.spinUp({
  gpuSkuId: gpus[0].id,
  name: 'gpu-host-01',
  imageName: 'ghcr.io/YOUR_ORG/your-gpu-app:latest',
  containerDiskGb: 50,
  ports: '8080/http,22/tcp',
  env: { MODEL_ID: 'your-model' },
})

const detail = await chassis.getInstance(instance.id)
console.log(detail.publicIp, detail.connection, detail.status)
// Point your clients at publicIp / published ports

await chassis.stop(instance.id)

Example: training job

Spin up a dedicated GPU, run your trainer on the instance, then stop billing.

const gpus = await chassis.listGpus()
const gpu = gpus.find((g) => g.displayName.includes('A100')) ?? gpus[0]

const instance = await chassis.spinUp({
  gpuSkuId: gpu.id,
  name: 'finetune-bert',
  gpuCount: 1,
  imageName: 'pytorch/pytorch:2.1.0-cuda11.8-cudnn8-devel',
  containerDiskGb: 80,
  ports: '8888/http,22/tcp',
})

console.log('running', instance.id, instance.status)
// SSH or open Jupyter on the instance, then run train.py

await chassis.stop(instance.id)

Example: multi-node cluster

const gpus = await chassis.listGpus()

const cluster = (await chassis.createCluster({
  name: 'dist-train',
  gpuSkuId: gpus[0].id,
  nodeCount: 4,
  gpusPerNode: 1,
  imageName: 'pytorch/pytorch:2.1.0-cuda11.8-cudnn8-devel',
})) as { id: string }

console.log(await chassis.getCluster(cluster.id))
// Nodes receive CHASSIS_CLUSTER_ID, CHASSIS_NODE_RANK, CHASSIS_NODE_COUNT

await chassis.stopCluster(cluster.id)
// await chassis.terminateCluster(cluster.id)

Example: serverless endpoint

const gpus = await chassis.listGpus()

const endpoint = await chassis.createEndpoint({
  name: 'text-infer',
  gpuSkuId: gpus[0].id,
  workersMin: 0,
  workersMax: 3,
})

const result = await chassis.runSync(endpoint.id, {
  input: {
    prompt: 'Summarize Chassis in one sentence.',
    max_tokens: 128,
  },
})
console.log(result)

const job = (await chassis.run(endpoint.id, {
  input: { prompt: 'hello' },
})) as { id: string }
const status = await chassis.getJob(endpoint.id, job.id)
console.log(status)

Instance options (createInstance / spinUp)

| Field | Required | Notes | |-------|----------|--------| | gpuSkuId | yes | Chassis SKU UUID from listGpus() | | name | yes | Instance name | | gpuCount | no | Default 1, max 8 (and SKU maxGpuCount) | | imageName | no | Container image | | containerDiskGb | no | Default 50 | | volumeGb | no | Ephemeral volume GB | | networkVolumeId | no | Chassis network volume UUID | | registryCredentialId | no | Chassis registry credential UUID | | cloudType | no | 'SECURE' | 'COMMUNITY' | | ports | no | e.g. "8080/http,22/tcp" | | env | no | String map passed into the container | | startAfterCreate | no | Default true (spinUp forces true) |

Methods

GPUs & instances

| Method | HTTP | |--------|------| | listGpus() | GET /gpus | | listInstances() | GET /instances | | createInstance(input) | POST /instances | | spinUp(input) | POST /instances (startAfterCreate: true) | | getInstance(id) | GET /instances/:id | | getInstanceLogs(id, tail?) | GET /instances/:id/logs | | updateInstance(id, input) | PATCH /instances/:id | | start(id) | POST /instances/:id/start | | stop(id) | POST /instances/:id/stop | | restart(id) | POST /instances/:id/restart | | terminate(id) | DELETE /instances/:id |

Clusters (multi-node)

| Method | HTTP | |--------|------| | listClusters() | GET /clusters | | createCluster(input) | POST /clusters | | getCluster(id) | GET /clusters/:id | | startCluster(id) | POST /clusters/:id/start | | stopCluster(id) | POST /clusters/:id/stop | | terminateCluster(id) | DELETE /clusters/:id |

Create a cluster with name, gpuSkuId, nodeCount (2–8), and optional gpusPerNode, imageName, networkVolumeId, ports. Nodes get CHASSIS_CLUSTER_ID, CHASSIS_NODE_RANK, and CHASSIS_NODE_COUNT env vars.

Templates, volumes, registries

| Method | HTTP | |--------|------| | listTemplates() / createTemplate(input) | GET / POST /templates | | updateTemplate(id, input) / deleteTemplate(id) | PATCH / DELETE /templates/:id | | listVolumes() / createVolume(input) | GET / POST /volumes | | updateVolume(id, input) | PATCH /volumes/:id | | listRegistries() / createRegistry(input) | GET / POST /registries |

Serverless endpoints

| Method | HTTP | |--------|------| | listEndpoints() | GET /endpoints | | createEndpoint(input) | POST /endpoints | | getEndpoint(id) / updateEndpoint(id, input) | GET / PATCH /endpoints/:id | | deleteEndpoint(id) | DELETE /endpoints/:id | | run(endpointId, body) | POST /endpoints/:id/run | | runSync(endpointId, body, waitMs?) | POST /endpoints/:id/runsync | | getEndpointHealth(id) | GET /endpoints/:id/health | | getJob(endpointId, jobId) | GET /endpoints/:id/jobs/:jobId | | cancelJob(endpointId, jobId) | POST /endpoints/:id/jobs/:jobId/cancel |

Serverless helpers:

const health = await chassis.getEndpointHealth(endpoint.id)

Create options also accept env, interruptible, and publicIp.

Responses use { data: … } / { error: "…" }. Failures raise ChassisError with a clear .message.

Docs

Human setup guide: https://chassis.okeymeta.com.ng/docs

License

MIT · OkeyMeta Ltd