BYOA指南

常见 问题

使用FanChat和你自己的API密钥所需了解的一切。

LLM供应商(文本)

选择供应商为你的角色提供能力。在设置中配置你的密钥。

OpenAI

OpenAI

Vision

Maker of GPT-4o and GPT-5. The market leader with the most versatile models.

模型
gpt-5.5gpt-5.4gpt-5.4-minigpt-5.4-nanogpt-5gpt-4.1gpt-4.1-minigpt-4ogpt-4o-minio3o4-mini
获取密钥

Create an account on OpenAI Platform, go to API Keys, click "Create new secret key".

Anthropic

Anthropic

Vision

Maker of Claude. Excellent for reasoning, code, and long conversations.

模型
claude-opus-4-7claude-opus-4-6claude-sonnet-4-6claude-sonnet-4-5claude-haiku-4-5-20251001claude-3-5-sonnet-20241022claude-3-5-haiku-20241022
获取密钥

Sign in to the Anthropic Console, go to Account > API Keys.

Google Gemini

Google Gemini

Vision

Google's Gemini family. Generous free tier, excellent quality/price ratio.

模型
gemini-3.1-pro-previewgemini-3.1-flash-lite-previewgemini-3-pro-previewgemini-3-flash-previewgemini-2.5-progemini-2.5-flashgemini-2.5-flash-litegemini-2.0-flash
获取密钥

Go to Google AI Studio, click "Get API Key". A Google account is enough.

Mistral

Mistral

Vision

Cutting-edge French AI. Fast, efficient models, excellent for French.

模型
mistral-large-2512mistral-small-2603mistral-large-latestmistral-small-latestmistral-medium-2508magistral-medium-latestmagistral-smallpixtral-large-latestcodestral-latest
获取密钥

Mistral Console > API Keys > "New API Key". Very competitive pricing.

Groq

Groq

Vision

Ultra-fast LPU inference. Ideal for instant responses with open-source models.

模型
meta-llama/llama-4-maverick-17b-128e-instructmeta-llama/llama-4-scout-17b-16e-instruct
获取密钥

Groq Console > API Keys > "Create API Key". Free tier available.

DeepSeek

DeepSeek

High-performance Chinese models at very low prices. DeepSeek-R1 excels at reasoning.

模型
deepseek-v4-prodeepseek-v4-flashdeepseek-chatdeepseek-reasoner
获取密钥

Create an account on DeepSeek Platform, go to API Keys.

Local (Gemma 4)

Local (Gemma 4)

免费

Local model running in your browser via WebGPU. No data sent to a server.

模型
gemma-4-e2b-it
获取密钥

No API key required. Initial model download (~1.2 GB), runs 100% offline.

图片供应商

用于生成角色图片和聊天中的图片。可选。

GPT Image (OpenAI)

GPT Image (OpenAI)

Image generation by OpenAI. gpt-image-2 with multi-reference and transparency support.

模型
gpt-image-2gpt-image-1
获取密钥

Same API key as for OpenAI text models.

Gemini Image (Google)

Gemini Image (Google)

Image generation by Google Gemini. Supports reference images and visual search.

模型
gemini-3.1-flash-image-previewgemini-2.5-flash-imagegemini-2.0-flash-preview-image-generation
获取密钥

Same API key as for Google text models.

FAQ

没有找到你的答案?

联系我们