AI models
OpenAI models in Quidget
Quidget is gradually migrating from the OpenAI Assistants API to the newer Responses API.
The Assistants API is deprecated and will be removed in August 2026.
The Responses API is the recommended replacement and is required to use GPT-5 models.
What to keep in mind
- After switching to the Responses API, model behavior may change slightly
- Existing prompts and instructions can behave a bit differently
- We recommend testing your agent after switching, especially if you rely on complex prompts or rules
Below is an overview of the currently supported OpenAI models in Quidget.
This guide helps you understand which model to choose, how it behaves, and which plans support it.
Models overview
| Model | Best for / Typical use cases | Supported plans |
|---|---|---|
| GPT-4o mini (default) | Fast, cost-efficient, and versatile model suitable for most chatbots FAQs, customer support, lead qualification, simple dialogs, high-traffic scenarios | All plans |
| GPT-5 nano | Ultra-fast and stable responses for short and simple outputs Button-based flows, order status, short factual answers, high-load environments | Pro, Pro+, Enterprise |
| GPT-5 mini | Balanced quality and speed with strong structured reasoning Rule-based support, knowledge base Q&A, multi-step responses | Pro, Pro+, Enterprise |
| GPT-5 | Handles complex logic and long contextual reasoning Advanced support cases, decision trees, B2B workflows | Pro+, Enterprise |
| GPT-5.1 | Improved accuracy and consistency over GPT-5 Mission-critical support, enterprise-grade workflows | Pro+, Enterprise |
| GPT-5.2 | Highest overall quality and reasoning performance Advanced enterprise use cases, complex logic, high-accuracy requirements | Pro+, Enterprise |
| GPT-5.4 nano | Lightweight, ultra-fast model optimized for efficiency High-volume requests, simple automations, low-latency scenarios | Enterprise |
| GPT-5.4 mini | Balanced performance with improved reasoning over nano models Scalable support flows, structured responses, moderate complexity tasks | Enterprise |
| GPT-4.1 nano | Lightweight model with configurable creativity Simple chatbots, fast responses, low-complexity use cases | All plans |
| GPT-4.1 mini | Better language quality with moderate cost General customer support, more natural tone | All plans |
| GPT-4.1 | Reliable general-purpose assistant Standard customer support, longer conversations | Pro, Pro+, Enterprise |
| GPT-4o | High language quality and natural tone of voice Premium chat experience, sales, onboarding, brand communication | Pro, Pro+, Enterprise |
Default model
- GPT-4o mini is the default model for newly created AI agents
- It provides the best balance of cost, performance, and response quality for most use cases.
How to choose the right model
General recommendations:
- 👉 GPT-4o mini — best default choice for ~80% of customer support chatbots, offering a strong balance of speed, cost, and quality
- 👉 GPT-5 mini — ideal when you need structured responses and consistent reasoning across similar queries
- 👉 GPT-5 / 5.1 / 5.2 — designed for complex logic, enterprise-grade workflows, and maximum accuracy
- 👉 GPT-4o / 4.1 — best when natural conversation, tone of voice, and human-like interaction are critical
- ❌ Deprecated models — use only to maintain existing or legacy setups
Deprecated models
The following models are deprecated and not recommended for new agents. The models listed below are deprecated and are no longer recommended for new agents.
They remain available solely for backward compatibility.
- GPT-4o (2024-11-20)
- GPT-4o mini (2024-07-18)
- GPT-4 Turbo
- GPT-3.5 Turbo
- GPT-3.5 Turbo (1106)
- GPT-3.5 Turbo (0125)







