Dedicated endpoints

Reserve model capacity by choosing a model and plan. Morph provisions and operates it; once ready, connect an OpenAI client using the endpoint’s URL and served model name.

Idle capacity remains allocated and billable. Endpoints do not automatically shrink with traffic or scale to zero.

Create an endpoint

CLI
Dashboard

Install the Morph CLI:

curl -fsSL https://morphllm.com/install.sh | bash
export PATH="$HOME/.local/bin:$PATH"

Create a dashboard API key for the endpoint’s account or organization, then set it in your terminal:

export MORPH_API_KEY="YOUR_MORPH_API_KEY"
morph dedicated models
morph dedicated create deepseek-v4-flash

models lists valid IDs; substitute one for deepseek-v4-flash if needed.Without a plan flag, create opens the browser plan picker. Select the owning account or organization there; the CLI key does not select browser billing. Choose capacity, review the agreement, and confirm.Alternatively, choose --price, --balanced, or --fast in the terminal:

morph dedicated create deepseek-v4-flash --balanced

The CLI shows the plan, requests confirmation, then uses saved payment details or opens hosted checkout. Use one purchase flow per endpoint.

Wait for readiness

Watch provisioning on the endpoint’s dashboard detail page, or list and inspect orders:

morph dedicated list
morph dedicated status ENDPOINT_ID

Use the ID from your purchase or endpoint list. Send requests only once state is ready. Provisioning must finish within a two-hour activation window that starts at payment; an order not ready by then is refunded automatically. Get connection details and lifecycle events as JSON:

morph dedicated status ENDPOINT_ID --json

After activation, endpoint.endpointUrl and endpoint.servedModelName identify your endpoint. Use this served name for requests, not the catalog ID.

Endpoint states

state is the field to act on. Every endpoint is in exactly one:

`state`	Meaning	Serves requests
`checkout_pending`	Waiting for payment.	No
`paid`, `capacity_held`, `provisioning`, `pr_open`, `merged`, `deploying`	Provisioning, in that order.	No
`ready`	Live.	Yes
`suspended`	Paused after a failed payment’s grace period; returns to `ready` once payment recovers.	No
`refunding`, `refunded`	Ended before activation: canceled, provisioning failed, no capacity, or the activation window elapsed. Refund in progress, then complete.	No
`failed`	Ended with an error that needs support. Rare.	No
`canceled`	Ended after activation, or checkout abandoned.	No

ready is the only serving state. The status response also has provisioningStatus (pending, pr_open, merged, deploying, ready, failed, or null), which tracks only the deployment job. Use it for diagnosis, not gating: it stays ready after an endpoint is suspended or canceled. To see why an order ended before activation, check the last lifecycle event in status --json or history: deployment.failed, capacity.unavailable, activation.timed_out, or cancellation.before_activation. refunded and canceled are final. Ended endpoints stay in your list indefinitely, and their endpointUrl and servedModelName remain populated, so always filter on state.

Discover endpoints from code

The CLI reads a public HTTP API you can call with the same key:

curl https://api.morphllm.com/api/v1/dedicated/endpoints \
  -H "Authorization: Bearer $MORPH_API_KEY"

Each listed endpoint includes state, endpointUrl, and servedModelName, so discovery needs no per-endpoint call. An organization key lists the organization’s endpoints; a personal key lists the user’s own. See the API reference for the list, status, and cancel routes, with example responses.

Send a request

Copy connection details from the dashboard or status response; set them alongside your key:

export MORPH_API_KEY="YOUR_MORPH_API_KEY"
export MORPH_ENDPOINT_URL="YOUR_ENDPOINT_URL"
export MORPH_ENDPOINT_MODEL="YOUR_SERVED_MODEL_NAME"

Supply the URL without /v1; these examples append it.

Python
JavaScript

Run pip install openai, then:

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["MORPH_API_KEY"],
    base_url=os.environ["MORPH_ENDPOINT_URL"].rstrip("/") + "/v1",
)

response = client.chat.completions.create(
    model=os.environ["MORPH_ENDPOINT_MODEL"],
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

Run npm install openai. Save as request.mjs; run node request.mjs:

import OpenAI from "openai";

const endpointUrl = process.env.MORPH_ENDPOINT_URL;
const model = process.env.MORPH_ENDPOINT_MODEL;
if (!endpointUrl || !model) {
  throw new Error("Set MORPH_ENDPOINT_URL and MORPH_ENDPOINT_MODEL from the endpoint status.");
}

const client = new OpenAI({
  apiKey: process.env.MORPH_API_KEY,
  baseURL: endpointUrl.replace(/\/$/, "") + "/v1",
});

const response = await client.chat.completions.create({
  model,
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Keys must belong to the endpoint’s account or organization. morph token new saves a CLI key; SDKs still need MORPH_API_KEY exported.

Monitor your endpoint

Open its dashboard:

morph dedicated dashboard ENDPOINT_ID

View provisioning, request counts, token usage, latency, and success metrics where available. Logs contain operational metadata, excluding prompts and responses.

morph dedicated logs ENDPOINT_ID
morph dedicated logs ENDPOINT_ID --errors
morph dedicated logs ENDPOINT_ID --follow
morph dedicated history ENDPOINT_ID

logs --follow polls until Ctrl+C; history shows lifecycle events, including provisioning. Logs are empty before activation because requests are not yet served.

Capacity and scaling

Lower traffic does not reduce reserved capacity or charges. The CLI and dashboard expose no autoscaling limits, scaling to zero, region selection, custom weights, or engine configuration. Morph manages serving and placement; contact us for other capacity arrangements. Endpoint access is scoped to your account, but hardware is not exclusive: idle hardware may serve other traffic. Use shared inference without reserving capacity.

Billing and cancellation

Hourly plans bill reserved GPU time from readiness, including idle time. Displayed token counts measure usage, not charges. Check your checkout agreement for rates and service terms. In Dedicated, select Cancel dedicated on the order or detail page and confirm. Organization cancellations require an admin. Alternatively:

morph dedicated stop ENDPOINT_ID

Canceling a purchased order before activation stops billing immediately, moves it to refunding, and refunds collected payment. Active endpoints follow their agreement: they stay ready, serving and billing, until the cancellation response’s effective date (cancelEffectiveAt in status), then move to canceled. Verify with morph dedicated status ENDPOINT_ID. Cancel accidental purchases before creating another endpoint.

Troubleshoot a request

Symptom	What to check
Still provisioning	Check `status` and `history`; wait for `state` `ready`.
Unauthorized	Use a valid key from the owning account or organization.
Model not found	Use the served model name, not the catalog ID.
Request fails	Check `logs --errors` and readiness.
Organization purchase or cancellation denied	Ask an organization admin.

For provisioning failures, send support the endpoint ID and history. Do not duplicate pending orders.

Get Started

Models

API Reference

Integrations

Create an endpoint

Wait for readiness

Endpoint states

Discover endpoints from code

Send a request

Monitor your endpoint

Capacity and scaling

Billing and cancellation

Troubleshoot a request

​Create an endpoint

​Wait for readiness

​Endpoint states

​Discover endpoints from code

​Send a request

​Monitor your endpoint

​Capacity and scaling

​Billing and cancellation

​Troubleshoot a request

Create an endpoint

Wait for readiness

Endpoint states

Discover endpoints from code

Send a request

Monitor your endpoint

Capacity and scaling

Billing and cancellation

Troubleshoot a request