OpenAI (GPT)
OpenAI provides the GPT and o-series models, widely used for general-purpose chat, code generation, reasoning, and image generation.
Getting an API Key
- Visit platform.openai.com/api-keys
- Sign in or create an account
- Generate a new API key (starts with
sk-...) - Paste the key into AI Supreme Council under Settings > AI Model > OpenAI
API keys are stored locally in your browser (localStorage) and are never included in shared bot URLs.
Supported Models
GPT Series (General Purpose)
| Model | Context Window | Max Output | Input Price | Output Price | Capabilities |
|---|---|---|---|---|---|
| GPT-5 | 400K | 128K | $1.25/MTok | $10.00/MTok | Vision, tools, reasoning, JSON mode, code |
| GPT-5 Mini | 400K | 128K | $0.25/MTok | $2.00/MTok | Vision, tools, reasoning, JSON mode, code |
| GPT-5 Nano | 400K | 128K | $0.05/MTok | $0.40/MTok | Vision, tools, JSON mode, code |
| GPT-4.1 | 1M | 32K | $2.00/MTok | $8.00/MTok | Vision, tools, JSON mode, code |
| GPT-4.1 Mini | 1M | 32K | $0.40/MTok | $1.60/MTok | Vision, tools, JSON mode, code |
| GPT-4.1 Nano | 1M | 32K | $0.10/MTok | $0.40/MTok | Vision, tools, JSON mode, code |
| GPT-4o | 128K | 16K | $2.50/MTok | $10.00/MTok | Vision, tools, JSON mode, code |
| GPT-4o Mini | 128K | 16K | $0.15/MTok | $0.60/MTok | Vision, tools, JSON mode, code |
Reasoning Models (o-series)
| Model | Context Window | Max Output | Input Price | Output Price | Capabilities |
|---|---|---|---|---|---|
| o3 | 200K | 100K | $2.00/MTok | $8.00/MTok | Vision, tools, reasoning |
| o3 Mini | 200K | 100K | $1.10/MTok | $4.40/MTok | Tools, reasoning, code |
| o4 Mini | 200K | 100K | $1.10/MTok | $4.40/MTok | Vision, tools, reasoning, code |
| o1 | 200K | 100K | $15.00/MTok | $60.00/MTok | Vision, tools, reasoning |
Prices are per million tokens (MTok). All models support cached input at reduced rates.
- GPT-5 is the latest flagship -- strong across all tasks with native reasoning built in.
- GPT-4.1 has a massive 1M token context window, ideal for large codebases and documents.
- o3 / o4-mini are dedicated reasoning models for complex math, logic, and analysis.
- GPT-4o Mini or GPT-4.1 Nano are great for cost-efficient everyday use.
Reasoning Models
The o-series models (o3, o3 Mini, o4 Mini, o1) are specialized for complex reasoning tasks. They think through problems step by step before answering.
To control reasoning depth, set the Reasoning Effort in the bot configuration panel:
| Setting | Behavior |
|---|---|
low | Minimal reasoning -- faster, cheaper |
medium | Balanced reasoning depth |
high | Deep reasoning -- best quality, slower |
Reasoning effort maps to the reasoning_effort parameter in the OpenAI API. Numeric values and max are mapped to high for OpenAI-compatible providers.
Vision Support
All GPT models and most o-series models support vision input. You can:
- Paste images directly into the chat input (Ctrl+V / Cmd+V)
- Upload images using the attachment button
- Drag and drop images into the chat area
GPT models can analyze images, read documents, describe visual content, and interpret charts and diagrams.
Tool Calling
OpenAI models have strong tool/function calling support. Define tools in your bot configuration, and the model will invoke them as structured function calls. AI Supreme Council displays tool calls as formatted blocks in the chat.
JSON Mode
GPT models support JSON mode, which constrains the output to valid JSON. To enable it, set the Response Format to json in the bot configuration panel. This is useful for structured data extraction, API response formatting, and integration workflows.
When JSON mode is enabled, you should include instructions in your system prompt about the expected JSON structure. The model will always return valid JSON, but the schema depends on your prompt.
Image Generation (DALL-E)
When the Image Generation capability is enabled in AI Supreme Council, OpenAI models can generate images via DALL-E. The app detects image generation requests in your messages and routes them to the appropriate API.
Pricing
OpenAI bills per input and output token, with rates varying by model. Refer to openai.com/pricing for current rates.
AI Supreme Council tracks token usage per provider in Settings > Usage.
Configuration
When creating a bot profile, select OpenAI as the provider and choose your preferred model. You can set a per-bot API key in the bot configuration panel to override the global key.
The OpenAI provider uses the Chat Completions API (api.openai.com/v1/chat/completions) with SSE streaming. Models prefixed with gpt-*, o1*, o3*, o4*, and gpt-5* are automatically routed to OpenAI.
Tips for Best Results
- Use GPT-4.1 for large contexts. Its 1M token window handles entire codebases in a single conversation.
- Use o3 or o4-mini for math and logic. Reasoning models significantly outperform standard GPT on complex analytical tasks.
- Enable JSON mode for structured output. Pair it with a clear schema in your system prompt for reliable structured responses.
- Start with cheaper models. Use GPT-4o Mini or GPT-5 Nano to prototype, then upgrade when you need more capability.