Passer au contenu principal

Documentation Index

Fetch the complete documentation index at: https://docs.aihubmix.com/llms.txt

Use this file to discover all available pages before exploring further.

Prise en charge de l’interface multi-fonctions de l’API Openai Responses ; les fonctions suivantes ont été lancées :
  • Text input : entrée texte
  • Image input : entrée image
  • Streaming : streaming
  • Web search : recherche web
  • Deep research : pour les tâches complexes d’analyse et de recherche
  • Reasoning : contrôle de la profondeur de raisonnement, prend en charge 4 niveaux (minimal / low / medium / high). Seule la série gpt-5 prend en charge minimal.
  • Verbosity : longueur de sortie ; la série gpt-5 prend en charge 3 niveaux (low / medium / high)
  • Functions : fonctions
  • Utilisation de l’outil image_generation : la création et la génération d’images sont facturées sous gpt-image-1.
  • Code Interpreter : permet aux modèles d’écrire et d’exécuter du Python pour résoudre des problèmes. reasoning.effort ‘minimal’ n’est pas pris en charge lors de l’utilisation du code interpreter avec gpt-5.
  • Remote MCP : appel à un serveur MCP distant
  • Computer Use : Computer Use

​
Utilisation (appel Python) :

Identique à la méthode d’appel officielle d’OpenAI, il suffit de remplacer api_key et base_url pour la redirection. La Chine continentale peut y accéder directement. 
client = OpenAI(
    api_key="AIHUBMIX_API_KEY", # Replace with the key you generated in AiHubMix
    base_url="https://aihubmix.com/v1"
)
  1. For inference models, the output inference summary can be controlled using the following parameter, with the detail richness of the summary ranked as detailed > auto > None, where auto provides the best balance.
"summary": "auto"
  1. Optional deep reasoning models: ‎⁠o3-deep-research⁠ and ‎⁠o4-mini-deep-research⁠, only supported on the ‎⁠responses⁠ endpoint.
  2. The gpt-5 series focuses on stable reasoning and consistent outputs, and no longer supports the temperature and top_p parameters for controlling randomness. If you need more freedom, you can try gpt-5-chat-latest, which supports temperature.
  3. Reasoning models (o series / gpt-5 series) have deprecated max_tokens. Please use max_completion_tokens for completions or max_output_tokens for responses to explicitly set the output token limit.
from openai import OpenAI

client = OpenAI(
    api_key="sk-***", # Replace with the key generated in your AIHubMix dashboard
    base_url="https://aihubmix.com/v1"
)

response = client.responses.create(
    model="gpt-5", # gpt-5, gpt-5-chat-latest, gpt-5-mini, gpt-5-nano
    input="Why is tarot divination effective? What are the underlying principles and transferable methods? Output format: Markdown", # GPT-5 does not output in Markdown format by default, so you need to explicitly specify it.
    reasoning={
        "effort": "minimal" # Reasoning depth – Controls how many reasoning tokens the model generates before producing a response. Value can be "minimal", "low", "medium" or "high". Default is "medium".
    },
    text={
        "verbosity": "low" # Output length – Verbosity determines how many output tokens are generated. Value can be "low", "medium", or "high". Models before GPT-5 defaulted to "medium" verbosity.
    },
    stream=True
)

for event in response:
  print(event)
Note:
  1. The latest codex-mini-latest does not support search.
  2. The Computer use feature requires integration with Playwright. It’s recommended to refer to the official repository.
Known issues:
  • Use cases are complex to invoke
  • Takes many screenshots, which is time-consuming and often unreliable
  • May trigger CAPTCHA or Cloudflare human verification, potentially leading to infinite loops
Dernière mise à jour : 2026-06-01