O3 Pro O1 Pro Support
functionOpenAI o1-pro & o3-pro Integration for Open Web UI
14 class Valves(BaseModel):15 NAME_PREFIX: str = Field(16 default="OpenAI: ",17 description="Prefix to be added before model names.",18 )19 BASE_URL: str = Field(20 default="https://api.openai.com/v1",21 description="Base URL for OpenAI API.",22 )23 API_KEYS: str = Field(24 default="",25 description="API keys for OpenAI, use comma to separate multiple keys", OpenAI o1-pro & o3-pro Integration for Open Web UI
Access OpenAI's most advanced reasoning models with comprehensive token usage tracking and cost analytics directly in Open Web UI.
🌟Features
- Premium Models: Access to o1-pro and o3-pro reasoning models
- Cost Tracking: Detailed per-message and cumulative conversation costs
- Token Analytics: Track input, output, and reasoning tokens separately
- Multi-Key Support: Round-robin load balancing across multiple API keys
- Smart Conversation Tracking: Maintains cost history throughout conversations
- Configurable Parameters: Adjust reasoning effort, output length, and more
📋Prerequisites
- Open Web UI installation
- OpenAI API key(s) with access to o1-pro and/or o3-pro models
- Basic understanding of token-based pricing
🚀Quick Start
- Install the function in Open Web UI
- Add your OpenAI API key(s) in settings
- Select "OpenAI: o1-pro" or "OpenAI: o3-pro" from the model dropdown
- Start chatting with cost tracking enabled
⚙️Configuration
| Setting | Description | Default |
|---|---|---|
| API_KEYS | OpenAI API key(s), comma-separated | (required) |
| THINKING_EFFORT | Reasoning intensity: low, medium, high | medium |
| SHOW_TOKEN_STATS | Display usage after responses | True |
| SHOW_CUMULATIVE_COST | Show conversation totals | True |
| MAX_OUTPUT_TOKENS | Maximum response length | 3200 |
| TIMEOUT_SECONDS | Request timeout | 600 |
💰Model Pricing
| Model | Input | Output | Best For |
|---|---|---|---|
| o1-pro | $150/M | $600/M | Complex logic, coding, math |
| o3-pro | $20/M | $80/M | Research, current events, general use |
📊Cost Tracking Example
📊 Token Usage (This Message):
- Input: 1,234 tokens
- Reasoning: 567 tokens
- Output: 890 tokens
- Total: 2,691 tokens
- Cost for this message: $0.1234
💰 Conversation Totals:
- Messages: 3
- Total Input: 3,456 tokens
- Total Output: 2,345 tokens
- Total Cost: $0.3456🎯Model Selection Guide
Choose o1-pro when you need:
- Advanced mathematical problem solving
- Complex code generation or debugging
- Multi-step logical reasoning
- Maximum reasoning capability
Choose o3-pro when you need:
- Cost-effective advanced reasoning
- General purpose queries
- Research and analysis tasks
⚡Tips for Cost Optimization
- Use
lowthinking effort for simple queries - Set appropriate
MAX_OUTPUT_TOKENSlimits - Use o3-pro instead of o1-pro when possible
- Monitor cumulative costs during long conversations
🔧Troubleshooting
No API keys configured
- Add your OpenAI API key in function settings
High costs
- Lower
THINKING_EFFORTsetting - Reduce
MAX_OUTPUT_TOKENS - Switch from o1-pro to o3-pro
Timeout errors
- Increase
TIMEOUT_SECONDS - Simplify complex queries
📝Notes
- Responses appear all at once (no streaming) to ensure accurate token counting
- Conversation costs reset when starting a new chat
- Web search is only available for o3-pro model
- Pricing subject to change per OpenAI's policies
🤝Support
For issues or suggestions, please reach out to me.
Version: 1.0.0
License: MIT License
Source Code: Github
Author: Karan Bansal · @karanb192
OpenAI o1-pro & o3-pro Integration for Open Web UI
Access OpenAI's most advanced reasoning models with comprehensive token usage tracking and cost analytics directly in Open Web UI.
🌟Features
- Premium Models: Access to o1-pro and o3-pro reasoning models
- Cost Tracking: Detailed per-message and cumulative conversation costs
- Token Analytics: Track input, output, and reasoning tokens separately
- Multi-Key Support: Round-robin load balancing across multiple API keys
- Smart Conversation Tracking: Maintains cost history throughout conversations
- Configurable Parameters: Adjust reasoning effort, output length, and more
📋Prerequisites
- Open Web UI installation
- OpenAI API key(s) with access to o1-pro and/or o3-pro models
- Basic understanding of token-based pricing
🚀Quick Start
- Install the function in Open Web UI
- Add your OpenAI API key(s) in settings
- Select "OpenAI: o1-pro" or "OpenAI: o3-pro" from the model dropdown
- Start chatting with cost tracking enabled
⚙️Configuration
| Setting | Description | Default |
|---|---|---|
| API_KEYS | OpenAI API key(s), comma-separated | (required) |
| THINKING_EFFORT | Reasoning intensity: low, medium, high | medium |
| SHOW_TOKEN_STATS | Display usage after responses | True |
| SHOW_CUMULATIVE_COST | Show conversation totals | True |
| MAX_OUTPUT_TOKENS | Maximum response length | 3200 |
| TIMEOUT_SECONDS | Request timeout | 600 |
💰Model Pricing
| Model | Input | Output | Best For |
|---|---|---|---|
| o1-pro | $150/M | $600/M | Complex logic, coding, math |
| o3-pro | $20/M | $80/M | Research, current events, general use |
📊Cost Tracking Example
📊 Token Usage (This Message):
- Input: 1,234 tokens
- Reasoning: 567 tokens
- Output: 890 tokens
- Total: 2,691 tokens
- Cost for this message: $0.1234
💰 Conversation Totals:
- Messages: 3
- Total Input: 3,456 tokens
- Total Output: 2,345 tokens
- Total Cost: $0.3456🎯Model Selection Guide
Choose o1-pro when you need:
- Advanced mathematical problem solving
- Complex code generation or debugging
- Multi-step logical reasoning
- Maximum reasoning capability
Choose o3-pro when you need:
- Cost-effective advanced reasoning
- General purpose queries
- Research and analysis tasks
⚡Tips for Cost Optimization
- Use
lowthinking effort for simple queries - Set appropriate
MAX_OUTPUT_TOKENSlimits - Use o3-pro instead of o1-pro when possible
- Monitor cumulative costs during long conversations
🔧Troubleshooting
No API keys configured
- Add your OpenAI API key in function settings
High costs
- Lower
THINKING_EFFORTsetting - Reduce
MAX_OUTPUT_TOKENS - Switch from o1-pro to o3-pro
Timeout errors
- Increase
TIMEOUT_SECONDS - Simplify complex queries
📝Notes
- Responses appear all at once (no streaming) to ensure accurate token counting
- Conversation costs reset when starting a new chat
- Web search is only available for o3-pro model
- Pricing subject to change per OpenAI's policies
🤝Support
For issues or suggestions, please reach out to me.
Version: 1.0.0
License: MIT License
Source Code: Github
Author: Karan Bansal · @karanb192

