Gebruik CometAPI POST /v1/chat/completions om gesprekken met meerdere berichten te verzenden en LLM-antwoorden te krijgen met Streaming-, temperature- en max_tokens-instellingen.
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key="<COMETAPI_KEY>",
)
completion = client.chat.completions.create(
model="gpt-5.4",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Hello!"},
],
)
print(completion.choices[0].message){
"id": "chatcmpl-DNA27oKtBUL8TmbGpBM3B3zhWgYfZ",
"object": "chat.completion",
"created": 1774412483,
"model": "gpt-4.1-nano-2025-04-14",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "Four",
"refusal": null,
"annotations": []
},
"logprobs": null,
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 29,
"completion_tokens": 2,
"total_tokens": 31,
"prompt_tokens_details": {
"cached_tokens": 0,
"audio_tokens": 0
},
"completion_tokens_details": {
"reasoning_tokens": 0,
"audio_tokens": 0,
"accepted_prediction_tokens": 0,
"rejected_prediction_tokens": 0
}
},
"service_tier": "default",
"system_fingerprint": "fp_490a4ad033"
}model te wijzigen.
base_url te wijzigen naar https://api.cometapi.com/v1.reasoning_effort is alleen van toepassing op reasoning-modellen (o-series, GPT-5.1+), en sommige modellen ondersteunen mogelijk logprobs of n > 1 niet.o1-pro) in plaats daarvan het endpoint responses.| Role | Description |
|---|---|
system | Stelt het gedrag en de persoonlijkheid van de assistant in. Wordt aan het begin van het gesprek geplaatst. |
developer | Vervangt system voor nieuwere modellen (o1+). Geeft instructies die het model moet volgen, ongeacht de input van de gebruiker. |
user | Berichten van de eindgebruiker. |
assistant | Eerdere modelantwoorden, gebruikt om de gespreksgeschiedenis te behouden. |
tool | Resultaten van tool-/function-calls. Moet tool_call_id bevatten dat overeenkomt met de oorspronkelijke tool call. |
developer de voorkeur boven system voor instructieberichten. Beide werken, maar developer zorgt voor sterker instructievolgend gedrag.content om multimodale berichten te versturen:
{
"role": "user",
"content": [
{"type": "text", "text": "Describe this image"},
{
"type": "image_url",
"image_url": {
"url": "https://example.com/image.png",
"detail": "high"
}
}
]
}
detail bepaalt de diepgang van de afbeeldingsanalyse:
low — sneller, gebruikt minder tokens (vaste kosten)high — gedetailleerde analyse, meer tokens verbruiktauto — het model beslist (standaard)stream is ingesteld op true, wordt de response geleverd als Server-Sent Events (SSE). Elke event bevat een chat.completion.chunk-object met incrementele content:
data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant"},"finish_reason":null}]}
data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}
data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"!"},"finish_reason":null}]}
data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}
data: [DONE]
stream_options.include_usage in op true. De usage-data verschijnt in de laatste chunk vóór [DONE].response_format:
{
"response_format": {
"type": "json_schema",
"json_schema": {
"name": "result",
"strict": true,
"schema": {
"type": "object",
"properties": {
"answer": {"type": "string"},
"confidence": {"type": "number"}
},
"required": ["answer", "confidence"],
"additionalProperties": false
}
}
}
}
json_schema) garandeert dat de output exact overeenkomt met je schema. JSON Object mode (json_object) garandeert alleen geldige JSON — de structuur wordt niet afgedwongen.{
"tools": [
{
"type": "function",
"function": {
"name": "get_weather",
"description": "Get current weather for a city",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "City name"}
},
"required": ["location"]
}
}
}
],
"tool_choice": "auto"
}
finish_reason: "tool_calls" en bevat de array message.tool_calls de functienaam en argumenten. Vervolgens voer je de functie uit en stuur je het resultaat terug als een tool-bericht met de overeenkomende tool_call_id.
| Veld | Beschrijving |
|---|---|
id | Unieke completion-identificatie (bijv. chatcmpl-abc123). |
object | Altijd chat.completion. |
model | Het model dat de response heeft gegenereerd (kan een versiesuffix bevatten). |
choices | Array met completion-keuzes (meestal 1, tenzij n > 1). |
choices[].message | Het responsebericht van de assistant met role, content en optioneel tool_calls. |
choices[].finish_reason | Waarom het model stopte: stop, length, tool_calls of content_filter. |
usage | Uitsplitsing van Token-verbruik: prompt_tokens, completion_tokens, total_tokens en gedetailleerde subaantallen. |
system_fingerprint | Fingerprint van de backendconfiguratie voor debuggen van reproduceerbaarheid. |
Parameterondersteuning bij verschillende providers
| Parameter | OpenAI GPT | Claude (via compat) | Gemini (via compat) |
|---|---|---|---|
temperature | 0–2 | 0–1 | 0–2 |
top_p | 0–1 | 0–1 | 0–1 |
n | 1–128 | alleen 1 | 1–8 |
stop | Tot 4 | Tot 4 | Tot 5 |
tools | ✅ | ✅ | ✅ |
response_format | ✅ | ✅ (json_schema) | ✅ |
logprobs | ✅ | ❌ | ❌ |
reasoning_effort | o-series, GPT-5.1+ | ❌ | ❌ (gebruik thinking voor Gemini native) |
max_tokens vs max_completion_tokens
max_tokens — De verouderde parameter. Werkt met de meeste modellen, maar is deprecated voor nieuwere OpenAI-modellen.max_completion_tokens — De aanbevolen parameter voor GPT-4.1, GPT-5 series en o-series modellen. Vereist voor reasoning-modellen omdat deze zowel output tokens als reasoning tokens omvat.system vs developer role
system — De traditionele instructierol. Werkt met alle modellen.developer — Geïntroduceerd met o1-modellen. Biedt sterkere instructie-opvolging voor nieuwere modellen. Valt terug op system-gedrag bij oudere modellen.developer voor nieuwe projecten die gericht zijn op GPT-4.1+ of o-series modellen.429 Too Many Requests tegenkomt, implementeer dan exponential backoff:
import time
import random
from openai import OpenAI, RateLimitError
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key="<COMETAPI_KEY>",
)
def chat_with_retry(messages, max_retries=3):
for i in range(max_retries):
try:
return client.chat.completions.create(
model="gpt-5.4",
messages=messages,
)
except RateLimitError:
if i < max_retries - 1:
wait_time = (2 ** i) + random.random()
time.sleep(wait_time)
else:
raise
messages-array:
messages = [
{"role": "developer", "content": "You are a helpful assistant."},
{"role": "user", "content": "What is Python?"},
{"role": "assistant", "content": "Python is a high-level programming language..."},
{"role": "user", "content": "What are its main advantages?"},
]
finish_reason?| Value | Betekenis |
|---|---|
stop | Natuurlijke voltooiing of een stop sequence bereikt. |
length | Limiet van max_tokens of max_completion_tokens bereikt. |
tool_calls | Het model heeft een of meer tool/function calls aangeroepen. |
content_filter | Output is gefilterd vanwege het contentbeleid. |
max_completion_tokens om de outputlengte te beperken.gpt-5.4-mini of gpt-5.4-nano voor eenvoudigere taken).usage.Bearer token authentication. Use your CometAPI key.
Model ID to use for this request. See the Models page for current options.
"gpt-4.1"
A list of messages forming the conversation. Each message has a role (system, user, assistant, or developer) and content (text string or multimodal content array).
Show child attributes
If true, partial response tokens are delivered incrementally via server-sent events (SSE). The stream ends with a data: [DONE] message.
Sampling temperature between 0 and 2. Higher values (e.g., 0.8) produce more random output; lower values (e.g., 0.2) make output more focused and deterministic. Recommended to adjust this or top_p, but not both.
0 <= x <= 2Nucleus sampling parameter. The model considers only the tokens whose cumulative probability reaches top_p. For example, 0.1 means only the top 10% probability tokens are considered. Recommended to adjust this or temperature, but not both.
0 <= x <= 1Number of completion choices to generate for each input message. Defaults to 1.
Up to 4 sequences where the API will stop generating further tokens. Can be a string or an array of strings.
Maximum number of tokens to generate in the completion. The total of input + output tokens is capped by the model's context length.
Number between -2.0 and 2.0. Positive values penalize tokens based on whether they have already appeared, encouraging the model to explore new topics.
-2 <= x <= 2Number between -2.0 and 2.0. Positive values penalize tokens proportionally to how often they have appeared, reducing verbatim repetition.
-2 <= x <= 2A JSON object mapping token IDs to bias values from -100 to 100. The bias is added to the model's logits before sampling. Values between -1 and 1 subtly adjust likelihood; -100 or 100 effectively ban or force selection of a token.
A unique identifier for your end-user. Helps with abuse detection and monitoring.
An upper bound for the number of tokens to generate, including visible output tokens and reasoning tokens. Use this instead of max_tokens for GPT-4.1+, GPT-5 series, and o-series models.
Specifies the output format. Use {"type": "json_object"} for JSON mode, or {"type": "json_schema", "json_schema": {...}} for strict structured output.
Show child attributes
A list of tools the model may call. Currently supports function type tools.
Show child attributes
Controls how the model selects tools. auto (default): model decides. none: no tools. required: must call a tool.
Whether to return log probabilities of the output tokens.
Number of most likely tokens to return at each position (0-20). Requires logprobs to be true.
0 <= x <= 20Controls the reasoning effort for o-series and GPT-5.1+ models.
low, medium, high Options for streaming. Only valid when stream is true.
Show child attributes
Specifies the processing tier.
auto, default, flex, priority Successful chat completion response.
Unique completion identifier.
"chatcmpl-abc123"
chat.completion "chat.completion"
Unix timestamp of creation.
1774412483
The model used (may include version suffix).
"gpt-5.4-2025-07-16"
Array of completion choices.
Show child attributes
Show child attributes
"default"
"fp_490a4ad033"
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key="<COMETAPI_KEY>",
)
completion = client.chat.completions.create(
model="gpt-5.4",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Hello!"},
],
)
print(completion.choices[0].message){
"id": "chatcmpl-DNA27oKtBUL8TmbGpBM3B3zhWgYfZ",
"object": "chat.completion",
"created": 1774412483,
"model": "gpt-4.1-nano-2025-04-14",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "Four",
"refusal": null,
"annotations": []
},
"logprobs": null,
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 29,
"completion_tokens": 2,
"total_tokens": 31,
"prompt_tokens_details": {
"cached_tokens": 0,
"audio_tokens": 0
},
"completion_tokens_details": {
"reasoning_tokens": 0,
"audio_tokens": 0,
"accepted_prediction_tokens": 0,
"rejected_prediction_tokens": 0
}
},
"service_tier": "default",
"system_fingerprint": "fp_490a4ad033"
}