OpenAI LLM Selector

Options

OpenAI chat model selection

Model selection

Whether all available models are listed or only selected compatible ones.

Available options:

• Default models: Shows default models for this model type.
• All models: Shows all models available for the provided API key. This includes models that may not be compatible with this specific endpoint, so it is the responsibility of the user to select a model that is compatible with this node.

Model

Select a chat-optimized OpenAI model to be used.

Available options:

• GPT-3.5 Turbo Instruct: Legacy instruct-style GPT-3.5 model; optimized for following single-turn instructions. Use only for backward compatibility; prefer GPT-4o mini. Docs
• Babbage-002: Legacy GPT-3.5 base model for simple tasks. Use only if migrating old applications; prefer GPT-4o mini for new projects. Docs
• Davinci-002: Legacy GPT-3.5 base model; more capable than Babbage-002 but deprecated. Use only for legacy compatibility; modern tasks should use GPT-4o or GPT-4.1. Docs
• GPT-3.5 Turbo: Legacy chat/text model; OpenAI recommends GPT-4o mini instead. Use only for backward compatibility. Docs
• GPT-3.5 Turbo (16k): Legacy 16k-context variant of GPT-3.5 Turbo. Use only for legacy apps needing longer context. Docs
• GPT-4: Legacy high-intelligence GPT model. Avoid for new projects; prefer GPT-4o or GPT-4.1. Docs
• GPT-4 (32k): Legacy 32k-context variant of GPT-4. Superseded by GPT-4o with larger context and lower cost.
• GPT-4 Turbo: Legacy GPT-4 variant; superseded by GPT-4o. Use GPT-4o instead for cheaper, faster performance. Docs
• GPT-4o: Fast, intelligent, flexible GPT model. Best general-purpose choice with multimodal support. Docs
• GPT-4o mini: Smaller 4o variant for efficiency; supports multimodality. Great for lightweight or cost-sensitive tasks. Docs
• GPT-4.1: Smartest non-reasoning model. Choose when you need the highest quality for complex instructions. Docs
• GPT-4.1 mini: Smaller, faster 4.1 model for cost-efficient tasks. Good balance between speed, cost, and quality. Docs
• GPT-4.1 nano: Ultra-lightweight GPT-4.1 model. Best when efficiency and low cost matter most. Docs
• GPT-5: Latest flagship model with advanced reasoning and coding. Best for cutting-edge apps and agentic tasks. Docs
• GPT-5 mini: Smaller, faster GPT-5 variant. Use when you want GPT-5 quality at lower cost. Docs
• GPT-5 nano: Ultra-small GPT-5 optimized for efficiency-critical tasks. Ideal for real-time or embedded use. Docs
• o1: Reasoning-first model that spends more time thinking before responding. Best for challenging reasoning problems. Docs
• o1 mini: Lightweight variant of o1 optimized for STEM reasoning. Good for math-heavy or technical tasks. Docs
• o3: Well-rounded reasoning model; excels at math, coding, and vision. Strong choice for multi-domain reasoning. Docs
• o3 mini: Smaller o3-class model balancing speed and accuracy. Good for reasoning tasks when cost is a concern. Docs
• o4 mini: Efficient reasoning model optimized for fast, affordable performance. Best for reasoning tasks with tight budgets. Docs

Specific model ID

Select from a list of all available OpenAI models. The model chosen has to be compatible with OpenAI's Chat API. This configuration will overwrite the default model configurations when set.

Model Parameters

Maximum response length (token)

The maximum number of tokens to generate.

This value, plus the token count of your prompt, cannot exceed the model's context length.

Temperature

Sampling temperature to use, between 0.0 and 2.0.

Higher values will lead to less deterministic answers.

Try 0.9 for more creative applications, and 0 for ones with a well-defined answer. It is generally recommended altering this, or Top-p, but not both.

Note: this setting is ignored for reasoning models like GPT-5 or o-series models.

Seed

Set the seed parameter to any integer of your choice to have (mostly) deterministic outputs. The default value of 0 means that no seed is specified.

If the seed and other model parameters are the same for each request, then responses will be mostly identical. There is a chance that responses will differ, due to the inherent non-determinism of OpenAI models.

Please note that this feature is in beta and only currently supported for gpt-4-1106-preview and gpt-3.5-turbo-1106 [1].

[1] OpenAI Cookbook

Number of concurrent requests

Maximum number of concurrent requests to LLMs that can be made, whether through API calls or to an inference server. Exceeding this limit may result in temporary restrictions on your access.

It is important to plan your usage according to the model provider's rate limits, and keep in mind that both software and hardware constraints can impact performance.

For OpenAI, please refer to the Limits page for the rate limits available to you.

Top-p sampling

An alternative to sampling with temperature, where the model considers the results of the tokens (words) with top_p probability mass. Hence, 0.1 means only the tokens comprising the top 10% probability mass are considered.

Views

This node has no views

Workflows

• No workflows found

Links

• No links available

Developers