AI Model Configuration#

MoryFlow supports 20+ AI providers—pick the one that works best for you.

Supported Providers#

Direct Providers#

Provider	Features	Recommended Models
OpenAI	Most popular, rich ecosystem	GPT-4o, GPT-4 Turbo
Anthropic	Claude series, great with long text	Claude 3.5 Sonnet
Google	Gemini series	Gemini Pro
xAI	Grok series	Grok-2
DeepSeek	Affordable, multilingual	DeepSeek V3
Moonshot	Chinese optimized	Moonshot-v1
Zhipu AI	Chinese company, strong Chinese	GLM-4

Aggregated Providers#

Provider	Features
OpenRouter	One key for multiple models, great for beginners
AIHubMix	Chinese aggregation service

Local Models#

Method	Features
Ollama	Open source, supports many models, fully offline

Configuring API Keys#

How to Configure#

1. Open Settings → Model Management
2. Find the provider you want
3. Click the configure button
4. Enter your API Key
5. Save

Where to Get API Keys#

OpenRouter (Recommended for Beginners)#

1. Go to openrouter.ai
2. Sign up
3. Go to Dashboard → Keys
4. Create a new API Key

OpenAI#

1. Go to platform.openai.com
2. Sign in
3. Go to API Keys page
4. Create new Secret Key

Anthropic#

1. Go to console.anthropic.com
2. Sign up/in
3. Go to API Keys
4. Create new key

DeepSeek#

1. Go to platform.deepseek.com
2. Sign up
3. Go to API Keys
4. Create new key

Selecting Models#

After configuring an API Key, you can select models from the top of the chat panel:

Model Recommendations#

Everyday Use#

Need	Recommended Model	Reason
All-around	Claude 3.5 Sonnet	Great quality, reasonable price
Fast responses	GPT-4o-mini	Quick, affordable
Chinese writing	DeepSeek V3	Strong Chinese, cost-effective

Professional Scenarios#

Scenario	Recommended Model	Reason
Coding	Claude 3.5 Sonnet	Best code capabilities
Long documents	Claude 3 Opus	200K context
Privacy sensitive	Ollama local models	Fully offline

Budget-Conscious#

Budget	Recommended Approach
Just trying	OpenRouter free credits
Low cost	DeepSeek V3 (~$0.07/10K words)
Completely free	Ollama local models

Adding Custom Models#

If your model isn't in the presets, you can add it:

1. Open Settings → Model Management
2. Click "Add Custom Model"
3. Fill in info:
   • Model name
   • Model ID
   • Provider
   • Context window size

Model Parameters#

For those who like to tinker:

Parameter	Description	Suggested Value
Temperature	Controls randomness	0.7 (default)
Max Tokens	Maximum output length	Adjust as needed
Top P	Nucleus sampling	1 (default)

Common Issues#

API Key Invalid#

1. Check you copied it completely (no extra spaces)
2. Confirm key hasn't expired
3. Confirm account balance is sufficient

Model Responds Slowly#

1. Switch to a faster model (e.g., GPT-4o-mini)
2. Check your network
3. Consider local models

Do I Need a VPN?#

• OpenAI, Anthropic, Google: May need depending on region
• DeepSeek, Zhipu, Moonshot: Generally accessible everywhere
• OpenRouter: Varies by region