Create a chat completion

OpenAI-compatible chat completions endpoint. Accepts the OpenAI chat completions request shape and routes to any catalogued model.

Also served at /v1/openai/chat/completions; both paths accept the same requests. This bare path matches the OpenAI SDK’s default base URL shape, so pointing the SDK at https://api-gateway.merge.dev/v1 works with no other changes.

Authentication

AuthorizationBearer

Your production key sent as a bearer token.

Request

This endpoint expects an object.

modelstringRequired

messageslist of objectsRequired

temperaturedouble or nullOptional

top_pdouble or nullOptional

ninteger or nullOptional

streambooleanOptionalDefaults to false

stopstring or list of strings or nullOptional

max_tokensinteger or nullOptional

max_completion_tokensinteger or nullOptional

presence_penaltydouble or nullOptional

frequency_penaltydouble or nullOptional

logit_biasmap from strings to doubles or nullOptional

userstring or nullOptional

toolslist of objects or nullOptional

tool_choicestring or map from strings to any or nullOptional

response_formatobject or nullOptional

OpenAI response format.

seedinteger or nullOptional

stream_optionsmap from strings to any or nullOptional

service_tierenum or nullOptional

Processing tier. ‘flex’ = discounted best-effort (slower, may be throttled); ‘priority’ = premium low-latency (not currently priced on any route, so it fails closed). Omit or ‘standard’ for normal processing. Allowed only on routes that price the requested tier, otherwise the request fails closed with 400.

Allowed values:

service_tier_fallbackbooleanOptionalDefaults to false

If the provider throttles the requested tier (429/503), retry once at the standard tier, billed at standard rates. The response service_tier shows the tier that actually served.

logprobsboolean or nullOptional

Return log probabilities of the output tokens.

top_logprobsinteger or nullOptional

Number of most likely tokens to return per position; requires logprobs.

thinkingmap from strings to any or nullOptional

Extended thinking settings for models that support it.

tagslist of objects or nullOptional

Key-value tags to attach to this request.

project_idstring or nullOptional

Project ID (UUID) to attribute this request to.

customerstring or nullOptional

Customer ID (UUID) to scope this request to, applying that customer’s routing policy, provider keys, budget, and usage attribution.

routing_policy_idstring or nullOptional

Routing policy ID for this request: an org-level policy, or a project or customer policy when the request carries that scope.

priority_orderlist of strings or objects or nullOptional

Ordered list of models to try, first success wins, with the same rules as priority_order on POST /v1/responses.

vendorstring or nullOptional

Restrict routing to one vendor (hosting platform).

vendorslist of strings or nullOptional

Ordered list of acceptable vendors; the first available one wins.

provider_optionsmap from strings to any or nullOptional

Provider-specific options, keyed by provider family.

prompt_cache_keystring or nullOptional

Prompt-cache routing hint forwarded to providers that cache automatically.

include_routing_metadatabooleanOptionalDefaults to false

Return routing metadata for this request.

include_guardrails_metadatabooleanOptionalDefaults to false

Return prompt-injection and DLP results in a top-level guardrails object.

Response

Successful Response

Errors

422

Unprocessable Entity Error