Skip to main content

​
POST /v1/completions

Generates a text completion for a given prompt string. This is the legacy completions endpoint, distinct from the chat completions endpoint. It is maintained for compatibility with older OpenAI integrations and providers that expose a complete endpoint.
For most use cases, prefer chat completions. Modern models such as GPT-4o and Claude are optimized for the chat format.

​
Request headers

​
x-portkey-provider
string
The provider to route the request to (e.g. openai). Required when not using a config.
​
x-portkey-api-key
string
Your provider API key.
​
x-portkey-config
string
A JSON config object or config ID that defines routing, fallbacks, retries, and more.
​
x-portkey-virtual-key
string
A virtual key ID from Portkey Cloud that maps to a stored provider credential.

​
Request body

​
model
string
required
The model to use for completion (e.g. gpt-3.5-turbo-instruct).
​
prompt
string | string[]
required
The prompt to generate a completion for. Accepts a string or an array of strings for batched input.
​
temperature
number
default:"1"
Sampling temperature between 0 and 2. Higher values produce more varied output.
​
max_tokens
integer
default:"100"
The maximum number of tokens to generate in the completion.
​
stream
boolean
default:"false"
When true, partial results are streamed as server-sent events and the stream ends with data: [DONE].
​
top_p
number
default:"1"
Nucleus sampling threshold. Only tokens within the top top_p probability mass are considered.
​
frequency_penalty
number
default:"0"
Number between -2.0 and 2.0. Positive values reduce repeated tokens.
​
presence_penalty
number
default:"0"
Number between -2.0 and 2.0. Positive values encourage new topics.
​
n
integer
default:"1"
The number of completions to generate for each prompt.
​
stop
string | string[]
One or more sequences at which to stop generating. The stop sequence is not included in the output.
​
logprobs
integer
Include the log probabilities of the logprobs most likely tokens at each position.
​
echo
boolean
default:"false"
When true, echo the prompt back in the response along with the completion.
​
best_of
integer
default:"1"
Generate best_of completions server-side and return the best. Incurs higher token cost.
​
logit_bias
object
A map of token IDs to bias values from -100 to 100. Adjusts the likelihood of specific tokens appearing.
​
seed
integer
A seed for deterministic sampling.
​
user
string
A unique identifier for the end user, used for monitoring.

​
Response

​
id
string
A unique identifier for this completion.
​
object
string
Always text_completion.
​
created
integer
Unix timestamp of when the completion was created.
​
model
string
The model used for the completion.
​
choices
object[]
An array of completion choices.
​
usage
object
Token usage for this request.

​
Code examples

curl http://localhost:8787/v1/completions \
  -H "Content-Type: application/json" \
  -H "x-portkey-provider: openai" \
  -H "x-portkey-api-key: $OPENAI_API_KEY" \
  -d '{
    "model": "gpt-3.5-turbo-instruct",
    "prompt": "The Portkey AI Gateway is",
    "max_tokens": 128,
    "temperature": 0.7
  }'

Build docs developers (and LLMs) love

Get started for freeTalk to us