OpenAI provides industry-leading AI models:
- GPT-5 Series - Latest generation language models
- o-series - Advanced reasoning capabilities
- Multimodal - Vision and audio understanding
- Codex - Specialized code generation
Available Models
GPT-5.1 Series
Enhanced GPT-5 models with improved performance.| Model | Input | Output | Best For |
|---|---|---|---|
| GPT-5.1 | $1.63/M | $13.00/M | General purpose, balanced performance |
| GPT-5.1 Chat | $1.63/M | $13.00/M | Chat applications, conversational AI |
| GPT-5.1 Codex | $1.63/M | $13.00/M | Code generation, development assistance |
| GPT-5.1 Codex Max | $1.63/M | $13.00/M | Advanced code generation, complex tasks |
| GPT-5.1 Codex Mini | $1.63/M | $13.00/M | Lightweight code tasks, fast responses |
GPT-5 Series
Latest generation models with enhanced capabilities.| Model | Input | Output | Best For |
|---|---|---|---|
| GPT-5 Pro | $19.50/M | $156.00/M | Complex analysis, research, premium quality |
| GPT-5 | $1.62/M | $13.00/M | General purpose, production apps |
| GPT-5 Chat Latest | $1.62/M | $13.00/M | Chat applications, conversational AI |
| GPT-5 Codex | $1.62/M | $13.00/M | Code generation, debugging, development |
| GPT-5 Mini | $0.33/M | $2.60/M | High-volume, cost-efficient tasks |
| GPT-5 Nano | $0.07/M | $0.52/M | Simple tasks, embeddings, classification |
GPT-4o Series
Multimodal models with vision capabilities.| Model | Input | Output | Best For |
|---|---|---|---|
| GPT-4o | $3.25/M | $13.00/M | Vision, multimodal tasks, production |
| GPT-4o (2024-11-20) | $3.25/M | $13.00/M | Latest stable 4o |
| GPT-4o (2024-05-13) | $3.25/M | $13.00/M | Original 4o release |
| GPT-4o Search Preview | $3.25/M | $13.00/M | Search-enhanced queries |
| GPT-4o Search Preview (2025-03-11) | $3.25/M | $13.00/M | Dated search preview |
| GPT-4o Mini | $0.20/M | $0.78/M | Fast, cost-efficient vision |
| GPT-4o Mini (2024-07-18) | $0.20/M | $0.78/M | Stable mini release |
| GPT-4o Mini Search Preview | $0.20/M | $0.78/M | Budget search tasks |
| GPT-4o Mini Search Preview (2025-03-11) | $0.20/M | $0.78/M | Dated mini search |
GPT-4.1 Series
Reliable, proven models for production.| Model | Input | Output | Best For |
|---|---|---|---|
| GPT-4.1 | $2.60/M | $10.40/M | Production applications, stability |
| GPT-4.1 (2025-04-14) | $2.60/M | $10.40/M | Dated stable version |
| GPT-4.1 Mini | $0.52/M | $2.08/M | Cost-efficient production |
| GPT-4.1 Mini (2025-04-14) | $0.52/M | $2.08/M | Dated mini version |
| GPT-4.1 Nano | $0.13/M | $0.52/M | High-volume, simple tasks |
| GPT-4.1 Nano (2025-04-14) | $0.13/M | $0.52/M | Dated nano version |
o-series (Reasoning Models)
Advanced reasoning models that “think” before responding.| Model | Input | Output | Best For |
|---|---|---|---|
| o1 Pro | $195.00/M | $780.00/M | Premium reasoning, complex research |
| o1 Pro (2025-03-19) | $195.00/M | $780.00/M | Dated pro version |
| o3 Deep Research | $130.00/M | $520.00/M | Extended multi-step research |
| o3 | $13.00/M | $52.00/M | Complex reasoning, math, code |
| o3 (2025-04-16) | $13.00/M | $52.00/M | Dated o3 version |
| o4-mini Deep Research | $26.00/M | $104.00/M | Cost-effective research |
| o1 | $19.50/M | $78.00/M | Advanced reasoning tasks |
| o1 (2024-12-17) | $19.50/M | $78.00/M | Dated o1 version |
| o3 Mini | $1.43/M | $5.72/M | Balanced reasoning tasks |
| o3 Mini (2025-01-31) | $1.43/M | $5.72/M | Dated mini version |
| o4-mini | $0.71/M | $2.86/M | Fast reasoning, lower cost |
| o4-mini (2025-04-16) | $0.71/M | $2.86/M | Dated o4-mini version |
Best Practices
Model Selection
Use GPT-5 for
- Latest capabilities
- Best overall performance
- Production applications
- Cutting-edge features
Use GPT-4o for
- Vision tasks
- Multimodal needs
- Consistent quality
- Stable production
Use o-series for
- Complex reasoning
- Math problems
- Code debugging
- Research tasks
Use Mini/Nano for
- High-volume requests
- Simple tasks
- Cost optimization
- Fast responses
Prompt Caching
Optimize costs with prompt caching:
- First request: Full cost ($1.25/M for GPT-5)
- Cached requests: 90% discount ($0.13/M)
- Cache duration: 5 minutes of inactivity
- Chatbots with large system prompts
- Document analysis with consistent context
- Multi-turn conversations
- RAG applications with static knowledge
Context Windows
OpenAI models support large context:- GPT-4o: Up to 128K tokens (~100K words)
- GPT-5: Up to 128K tokens (~100K words)
- o-series: Up to 128K tokens (includes reasoning tokens)
Support
Need help with OpenAI integration?OpenAI Documentation
Official OpenAI documentation
OpenAI Pricing
Official pricing details
Splox Docs
Browse our guides
Community
Get help from the community