> For the complete documentation index, see [llms.txt](https://docs.stammer.ai/stammer.ai-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.stammer.ai/stammer.ai-docs/chat-ai-agents/general-settings/model-version.md).

# Model Version

### What is a Model?

A model is the AI engine behind your agent. Each model has different speed, accuracy, and cost trade-offs.

### What's the Difference?

Here's a quick breakdown of the current models you can use:

#### OpenAI Models

**GPT-4o (2024-11-20)**

* Best-in-class performance and reasoning
* Ideal for advanced use cases, voice agents, and long replies
* $0.0109 /avg response

**GPT-4o-mini**

* Fast, efficient, and highly affordable
* Great for general tasks, FAQs, and quick replies
* $0.0007 /avg response

**GPT-4O (2024-08-06)**

* Previous version of GPT-4o with solid performance
* Reliable for production workloads
* $0.0109 /avg response

**GPT-4.1**

* Strong reasoning and accuracy
* Works well for creative tasks or agents needing nuance
* $0.0088 /avg response

**GPT-4.1-mini**

* Balanced power and cost
* Smart choice for mid-volume use
* $0.0017 /avg response

**GPT-4.1-nano**

* Ultra lightweight, fast, and budget-friendly
* Best for simple Q\&A and high-volume bots
* $0.0004 /avg response

**GPT-5**

* OpenAI's most advanced model with unified reasoning
* Exceptional for coding, complex problem-solving, and expert-level responses
* Built-in smart routing between fast and deep-thinking modes
* $0.0055 /avg response

**GPT-5-mini**

* Cost-effective variant with strong capabilities
* Optimized for speed while maintaining quality
* Great for lighter applications with good reasoning
* $0.0011 /avg response

**GPT-5-nano**

* Ultra-fast, ultra-low-latency model
* Optimized for real-time and embedded applications
* Perfect for high-throughput, speed-critical tasks
* $0.0002 /avg response

#### Anthropic (Claude) Models

**CLAUDE-3.7 Sonnet**

* First hybrid reasoning model combining instant and extended thinking
* State-of-the-art for coding, agentic tasks, and complex workflows
* Exceptional at long context understanding and structured outputs
* Can think through problems step-by-step when needed
* $0.0131 /avg response

**CLAUDE-3.5 Haiku**

* Fast, light, and efficient
* Great choice for speed-focused use cases
* Good balance of quality and performance
* $0.0011 /avg response

#### xAI Models

**GROK-2-1212**

* Enhanced accuracy and instruction-following
* Strong multilingual capabilities
* More personality and conversational style
* Good mix of speed, accuracy, and creativity
* $0.0088 /avg response

***

**Key takeaways:**

* **For complex reasoning & coding**: GPT-5, Claude 3.7 Sonnet
* **For best value**: GPT-4o-mini, GPT-4.1-nano, GPT-5-nano, Claude 3.5 Haiku
* **For balanced performance**: GPT-4.1, GPT-5-mini, Grok-2-1212
* **For production reliability**: GPT-4o (2024-11-20), GPT-4.1

{% embed url="<https://youtu.be/BpF_3IoRV18>" %}