Models
OpenAI Models
Detailed overview and use cases of OpenAI models integrated with Lyzr
Basic Overview
Lyzr integrates multiple OpenAI models designed to suit various AI workloads. These models differ in cost, latency, reasoning abilities, and input types (text, vision, etc.), making it easy to choose the right model based on your use case.GPT-4o
GPT-4o
Multi-modal flagship model that supports text, image, and audio inputs. Offers GPT-4-level intelligence with lower latency and cost.Use Cases:
- Vision + text-based RAG agents
- Multi-modal chat assistants
- Chart/image understanding
- Vision + text
- Lower latency than GPT-4-turbo
- High reasoning quality
GPT-4o-mini
GPT-4o-mini
A smaller and faster variant of GPT-4o. Ideal for scalable applications where speed and cost are a priority.Use Cases:
- High-traffic chatbots
- Real-time feedback systems
- Conversational UI at scale
- Lightweight
- Cost-efficient
- Reasonably accurate
GPT-4.1
GPT-4.1
One of the most advanced models available with enhanced reasoning, better tool usage, and fewer hallucinations.Use Cases:
- RAG pipelines with structured data
- Analytical agents
- Code generation and debugging
- Available in Studio
- Enhanced reasoning
- Great for logic-heavy tasks
o1 / o1-preview / o1-mini
o1 / o1-preview / o1-mini
Experimental OpenAI models optimized for different types of applications. Suitable for early testing and feature validation.Use Cases:
- Prototype agents
- Research and testing
- Tunable agent configurations
- Experimental variants
- Different sizes (mini/preview)
- Fast iterations
o3-mini
o3-mini
Lightweight and coordination-aware model optimized for agent-based frameworks. Great for managing tasks in workflows.Use Cases:
- Task-based agents
- DAG-based agent orchestration
- Sub-agent task execution
- Fast
- Coordination-efficient
- Excellent for planning workflows
⚙️ All OpenAI models listed above are pre-integrated in Lyzr and available for immediate use in Lyzr Studio[How to use?]