Completions

POST /v1/completions

Plain text completion — the right endpoint for base models without a chat template. Streaming supported.

Request body

Field	Type	Default	Notes
model	string	(the deployment’s model)	Optional; must match if set
prompt	string | string[]	required
stream	bool	false
temperature / top_p	float	engine default
max_tokens	int	engine default
frequency_penalty / presence_penalty	float	engine default
stop	string | string[]	—
n	int	1

Other OpenAI fields (best_of, echo, logprobs, suffix, seed, …) pass through to the engine.

Example

curl

curl https://<your-deployment>.inference.meshive.ai/v1/completions \
  -H "Authorization: Bearer $MESHIVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "The capital of France is", "max_tokens": 16}'

Python

resp = client.completions.create(
    model="",
    prompt="The capital of France is",
    max_tokens=16,
)
print(resp.choices[0].text)

Rate limiting, timeouts, streaming error events, and status codes are identical to Chat completions.