OpenAI

To use the OpenAI API, set the OPENAI_API_KEY environment variable, specify via apiKey field in the configuration file or pass the API key as an argument to the constructor.

Example:

export OPENAI_API_KEY=your_api_key_here

The OpenAI provider supports the following model formats:

• openai:chat - defaults to gpt-3.5-turbo
• openai:completion - defaults to text-davinci-003
• openai:<model name> - uses a specific model name (mapped automatically to chat or completion endpoint)
• openai:chat:<model name> - uses any model name against the /v1/chat/completions endpoint
• openai:chat:ft:gpt-3.5-turbo-0613:company-name:ID - example of a fine-tuned chat completion model
• openai:completion:<model name> - uses any model name against the /v1/completions endpoint
• openai:embeddings:<model name> - uses any model name against the /v1/embeddings endpoint
• openai:assistant:<assistant id> - use an assistant

The openai:<endpoint>:<model name> construction is useful if OpenAI releases a new model, or if you have a custom model. For example, if OpenAI releases gpt-5 chat completion, you could begin using it immediately with openai:chat:gpt-5.

The OpenAI provider supports a handful of configuration options, such as temperature, functions, and tools, which can be used to customize the behavior of the model like so:

providers:
  - id: openai:gpt-3.5-turbo
    config:
      temperature: 0
      max_tokens: 1024

Formatting chat messages

The prompt file supports a message in OpenAI's JSON prompt format. This allows you to set multiple messages including the system prompt. For example:

[
  { "role": "system", "content": "You are a helpful assistant." },
  { "role": "user", "content": "Who won the world series in {{ year }}?" }
]

Equivalent yaml is also supported:

- role: system
  content: You are a helpful assistant.
- role: user
  content: Who won the world series in {{ year }}?

Configuring parameters

The providers list takes a config key that allows you to set parameters like temperature, max_tokens, and others. For example:

providers:
  - id: openai:gpt-3.5-turbo-0613
    config:
      temperature: 0
      max_tokens: 128
      apiKey: sk-abc123

Supported parameters include:

Parameter	Description
temperature	Controls the randomness of the AI's output. Higher values (close to 1) make the output more random, while lower values (close to 0) make it more deterministic.
max_tokens	Controls the maximum length of the output in tokens.
top_p	Controls the nucleus sampling, a method that helps control the randomness of the AI's output.
frequency_penalty	Applies a penalty to frequent tokens, making them less likely to appear in the output.
presence_penalty	Applies a penalty to new tokens (tokens that haven't appeared in the input), making them less likely to appear in the output.
best_of	Controls the number of alternative outputs to generate and select from.
functions	Allows you to define custom functions. Each function should be an object with a name, optional description, and parameters.
function_call	Controls whether the AI should call functions. Can be either 'none', 'auto', or an object with a name that specifies the function to call.
tools	Allows you to define custom tools. See OpenAI Tools documentation
tool_choice	Controls whether the AI should use a tool. See OpenAI Tools documentation
stop	Defines a list of tokens that signal the end of the output.
stop	Defines a list of tokens that signal the end of the output.
response_format	Response format restrictions.
seed	Seed used for deterministic output. Defaults to 0
apiKey	Your OpenAI API key, equivalent to OPENAI_API_KEY environment variable
apiKeyEnvar	An environment variable that contains the API key
apiHost	The hostname of the OpenAI API, please also read OPENAI_API_HOST below.
apiBaseUrl	The base URL of the OpenAI API, please also read OPENAI_BASE_URL below.
organization	Your OpenAI organization key.
headers	Additional headers to include in the request.

Here are the type declarations of config parameters:

interface OpenAiConfig {
  // Completion parameters
  temperature?: number;
  max_tokens?: number;
  top_p?: number;
  frequency_penalty?: number;
  presence_penalty?: number;
  best_of?: number;
  functions?: {
    name: string;
    description?: string;
    parameters: any;
  }[];
  function_call?: 'none' | 'auto' | { name: string };
  tools?: {
    type: string;
    function: {
      name: string;
      description?: string;
      parameters: any;
    };
  }[];
  tool_choice?: 'none' | 'auto' | { type: 'function'; function?: { name: string } } | { type: 'file_search' };
  stop?: string[];
  response_format?: { type: string };
  seed?: number;

  // General OpenAI parameters
  apiKey?: string;
  apiKeyEnvar?: string;
  apiHost?: string;
  apiBaseUrl?: string;
  organization?: string;
  headers?: { [key: string]: string };
}

Chat conversations

The OpenAI provider supports full "multishot" chat conversations, including multiple assistant, user, and system prompts.

The most straightforward way to do this is by creating a list of {role, content} objects. Here's an example:

prompts: [prompt.json]

providers: [openai:gpt-3.5-turbo]

tests:
  - vars:
      messages:
        - role: system
          content: Respond as a pirate
        - role: user
          content: Who founded Facebook?
        - role: assistant
          content: Mark Zuckerberg
        - role: user
          content: Did he found any other companies?

Then the prompt itself is just a JSON dump of messages:

{{ messages | dump }}

Simplified chat markup

Alternatively, you may prefer to specify a list of role: message, like this:

tests:
  - vars:
      messages:
        - user: Who founded Facebook?
        - assistant: Mark Zuckerberg
        - user: Did he found any other companies?

This simplifies the config, but we need to work some magic in the prompt template:

[
{% for message in messages %}
  {% set outer_loop = loop %}
  {% for role, content in message %}
  {
    "role": "{{ role }}",
    "content": "{{ content }}"
  }{% if not (loop.last and outer_loop.last) %},{% endif %}
  {% endfor %}
{% endfor %}
]

Creating a conversation history fixture

Using nunjucks templates, we can combine multiple chat messages. Here's an example in which the previous conversation is a fixture for all tests. Each case tests a different follow-up message:

# Set up the conversation history
defaultTest:
  vars:
    system_message: Answer concisely
    messages:
      - user: Who founded Facebook?
      - assistant: Mark Zuckerberg
      - user: What's his favorite food?
      - assistant: Pizza

# Test multiple follow-ups
tests:
  - vars:
      question: Did he create any other companies?
  - vars:
      question: What is his role at Internet.org?
  - vars:
      question: Will he let me borrow $5?

In the prompt template, we construct the conversation history followed by a user message containing the question:

[
  {
    "role": "system",
    "content": {{ system_message | dump }}
  },
  {% for message in messages %}
    {% for role, content in message %}
      {
        "role": "{{ role }}",
        "content": {{ content | dump }}
      },
    {% endfor %}
  {% endfor %}
  {
    "role": "user",
    "content": {{ question | dump }}
  }
]

info

Variables containing multiple lines and quotes are automatically escaped in JSON prompt files.

If the file is not valid JSON (such as in the case above, due to the nunjucks {% for %} loops), use the built-in nunjucks filter dump to stringify the object as JSON.

Using the _conversation variable

A built-in _conversation variable contains the full prompt and previous turns of a conversation. Use it to reference previous outputs and test an ongoing chat conversation.

The _conversation variable has the following type signature:

type Completion = {
  prompt: string | object;
  input: string;
  output: string;
};

type Conversation = Completion[];

In most cases, you'll loop through the _conversation variable and use each Completion object.

Use completion.prompt to reference the previous conversation. For example, to get the number of messages in a chat-formatted prompt:

{{ completion.prompt.length }}

Or to get the first message in the conversation:

{{ completion.prompt[0] }}

Use completion.input as a shortcut to get the last user message. In a chat-formatted prompt, input is set to the last user message, equivalent to completion.prompt[completion.prompt.length - 1].content.

Here's an example test config. Note how each question assumes context from the previous output:

tests:
  - vars:
      question: Who founded Facebook?
  - vars:
      question: Where does he live?
  - vars:
      question: Which state is that in?

Here is the corresponding prompt:

[
  {% for completion in _conversation %}
    {
      "role": "user",
      "content": "{{ completion.input }}"
    },
    {
      "role": "assistant",
      "content": "{{ completion.output }}"
    },
  {% endfor %}
  {
    "role": "user",
    "content": "{{ question }}"
  }
]

The prompt inserts the previous conversation into the test case, creating a full turn-by-turn conversation:

Try it yourself by using the full example config.

info

When the _conversation variable is present, the eval will run single-threaded (concurrency of 1).

Including JSON in prompt content

In some cases, you may want to send JSON within the OpenAI content field. In order to do this, you must ensure that the JSON is properly escaped.

Here's an example that prompts OpenAI with a JSON object of the structure {query: string, history: {reply: string}[]}. It first constructs this JSON object as the input variable. Then, it includes input in the prompt with proper JSON escaping:

{% set input %}
{
    "query": "{{ query }}",
    "history": [
      {% for completion in _conversation %}
        {"reply": "{{ completion.output }}"} {% if not loop.last %},{% endif %}
      {% endfor %}
    ]
}
{% endset %}

[{
  "role": "user",
  "content": {{ input | trim | dump }}
}]

Here's the associated config:

prompts: [prompt.json]
providers: [openai:gpt-3.5-turbo-0613]
tests:
  - vars:
      query: how you doing
  - vars:
      query: need help with my passport

This has the effect of including the conversation history within the prompt content. Here's what's sent to OpenAI for the second test case:

[
  {
    "role": "user",
    "content": "{\n    \"query\": \"how you doing\",\n    \"history\": [\n      \n    ]\n}"
  }
]

Images / gpt-4-vision

You can include images in the prompt by using content blocks.

See OpenAI vision example.

Using tools and functions

OpenAI tools and functions are supported. See OpenAI tools example and OpenAI functions example.

Using tools

To set tools on an OpenAI provider, use the provider's config key. Add your function definitions under this key.

prompts: [prompt.txt]
providers:
  - 'openai:chat:gpt-3.5-turbo-0613':
    config:
      tools: [
        {
        "type": "function",
          "function": {
            "name": "get_current_weather",
            "description": "Get the current weather in a given location",
            "parameters": {
              "type": "object",
                "properties": {
                  "location": {
                    "type": "string",
                      "description": "The city and state, e.g. San Francisco, CA"
                    },
                    "unit": {
                      "type": "string",
                      "enum": ["celsius", "fahrenheit"]
                    }
                  },
              "required": ["location"]
            }
          }
        }
      ]
      tool_choice: 'auto'

tests:
   - vars:
        city: Boston
     assert:
        - type: is-json
        - type: is-valid-openai-tools-call
        - type: javascript
          value: output[0].function.name === 'get_current_weather'
        - type: javascript
          value: JSON.parse(output[0].function.arguments).location === 'Boston, MA'

   - vars:
        city: New York
# ...

Sometimes OpenAI function calls don't match tools schemas. Use is-valid-openai-tools-call or is-valid-openai-tools-call assertions to enforce an exact schema match between tools and the function definition.

To further test tools definitions, you can use the javascript assertion and/or transform directives. For example:

tests:
  - vars:
      city: Boston
    assert:
      - type: is-json
      - type: is-valid-openai-tools-call
      - type: javascript
        value: output[0].function.name === 'get_current_weather'
      - type: javascript
        value: JSON.parse(output[0].function.arguments).location === 'Boston, MA'

  - vars:
      city: New York
      # transform returns only the 'name' property
    transform: output[0].function.name
    assert:
      - type: is-json
      - type: similar
        value: NYC

Using functions

functions and function_call is deprecated in favor of tools and tool_choice, see detail in OpenAI API reference.

In addition, you can use functions to define custom functions. Each function should be an object with a name, optional description, and parameters. For example:

prompts: [prompt.txt]
providers:
  - id: openai:chat:gpt-3.5-turbo-0613
    config:
      functions:
        [
          {
            'name': 'get_current_weather',
            'description': 'Get the current weather in a given location',
            'parameters':
              {
                'type': 'object',
                'properties':
                  {
                    'location':
                      {
                        'type': 'string',
                        'description': 'The city and state, e.g. San Francisco, CA',
                      },
                    'unit': { 'type': 'string', 'enum': ['celsius', 'fahrenheit'] },
                  },
                'required': ['location'],
              },
          },
        ]
tests:
  - vars:
      city: Boston
    assert:
      - type: is-valid-openai-function-call
  - vars:
      city: New York
  # ...

Sometimes OpenAI function calls don't match functions schemas. Use is-valid-openai-function-call assertions to enforce an exact schema match between function calls and the function definition.

To further test function call definitions, you can use the javascript assertion and/or transform directives. For example:

tests:
  - vars:
      city: Boston
    assert:
      - type: is-valid-openai-function-call
      - type: javascript
        value: output.name === 'get_current_weather'
      - type: javascript
        value: JSON.parse(output.arguments).location === 'Boston, MA'

  - vars:
      city: New York
    # transform returns only the 'name' property for this testcase
    transform: output.name
    assert:
      - type: is-json
      - type: similar
        value: NYC

Loading tools/functions from a file

Instead of duplicating function definitions across multiple configurations, you can reference an external YAML (or JSON) file that contains your functions. This allows you to maintain a single source of truth for your functions, which is particularly useful if you have multiple versions or regular changes to definitions.

To load your functions from a file, specify the file path in your provider configuration like so:

providers:
  - file://./path/to/provider_with_function.yaml

Here's an example of how your provider_with_function.yaml might look:

id: openai:chat:gpt-3.5-turbo-0613
config:
  functions:
    - name: get_current_weather
      description: Get the current weather in a given location
      parameters:
        type: object
        properties:
          location:
            type: string
            description: The city and state, e.g. San Francisco, CA
          unit:
            type: string
            enum:
              - celsius
              - fahrenheit
            description: The unit in which to return the temperature
        required:
          - location

Supported environment variables

These OpenAI-related environment variables are supported:

Variable	Description
OPENAI_TEMPERATURE	Temperature model parameter, defaults to 0.
OPENAI_MAX_TOKENS	Max_tokens model parameter, defaults to 1024.
OPENAI_API_HOST	The hostname to use (useful if you're using an API proxy). Takes priority over OPENAI_BASE_URL.
OPENAI_BASE_URL	The base URL (protocol + hostname + port) to use, this is a more general option than OPENAI_API_HOST.
OPENAI_API_KEY	OpenAI API key.
OPENAI_ORGANIZATION	The OpenAI organization key to use.
PROMPTFOO_REQUIRE_JSON_PROMPTS	By default the chat completion provider will wrap non-JSON messages in a single user message. Setting this envar to true disables that behavior.
PROMPTFOO_DELAY_MS	Number of milliseconds to delay between API calls. Useful if you are hitting OpenAI rate limits (defaults to 0).
PROMPTFOO_REQUEST_BACKOFF_MS	Base number of milliseconds to backoff and retry if a request fails (defaults to 5000).

Evaluating assistants

To test out an Assistant via OpenAI's Assistants API, first create an Assistant in the API playground.

Set functions, code interpreter, and files for retrieval as necessary.

Then, include the assistant in your config:

prompts:
  - 'Write a tweet about {{topic}}'
providers:
  - openai:assistant:asst_fEhNN3MClMamLfKLkIaoIpgZ
tests:
  - vars:
      topic: bananas
  # ...

Code interpreter, function calls, and retrievals will be included in the output alongside chat messages. Note that the evaluator creates a new thread for each eval.

The following properties can be overwritten in provider config:

• model - OpenAI model to use
• instructions - System prompt
• tools - Enabled tools
• thread.messages - A list of message objects that the thread is created with.
• temperature - Temperature for the model
• toolChoice - Controls whether the AI should use a tool

Here's an example of a more detailed config:

prompts:
  - 'Write a tweet about {{topic}}'
providers:
  - id: openai:assistant:asst_fEhNN3MClMamLfKLkIaoIpgZ
    config:
      model: gpt-4-1106-preview
      instructions: "You always speak like a pirate"
      temperature: 0.2
      toolChoice:
        type: file_search
      tools:
        - type: code_interpreter
        - type: file_search
      thread:
        messages:
          - role: user
            content: "Hello world"
          - role: assistant
            content: "Greetings from the high seas"
tests:
  - vars:
      topic: bananas
  # ...

Automatically handling function tool calls

You can specify JavaScript callbacks that are automatically called to create the output of a function tool call.

This requires definining your config in a JavaScript file instead of YAML.

module.exports = /** @type {import('promptfoo').TestSuiteConfig} */ ({
  prompts: 'Please add the following numbers together: {{a}} and {{b}}',
  providers: [
    {
      id: 'openai:assistant:asst_fEhNN3MClMamLfKLkIaoIpgZ',
      config:
        /** @type {InstanceType<import('promptfoo')["providers"]["OpenAiAssistantProvider"]>["config"]} */ ({
          model: 'gpt-4-1106-preview',
          instructions: 'You can add two numbers together using the `addNumbers` tool',
          tools: [
            {
              type: 'function',
              function: {
                name: 'addNumbers',
                description: 'Add two numbers together',
                parameters: {
                  type: 'object',
                  properties: {
                    a: { type: 'number' },
                    b: { type: 'number' },
                  },
                  required: ['a', 'b'],
                },
              },
            },
          ],
          /**
           * Map of function tool names to function callback.
           */
          functionToolCallbacks: {
            // this function should accept a string, and return a string
            // or a `Promise<string>`.
            addNumbers: (parametersJsonString) => {
              const { a, b } = JSON.parse(parametersJsonString);
              return JSON.stringify(a + b);
            },
          },
        }),
    },
  ],
  tests: [
    {
      vars: { a: 5, b: 6 },
    },
  ],
});

Troubleshooting

OpenAI rate limits

There are a few things you can do if you encounter OpenAI rate limits (most commonly with GPT-4):

1. Reduce concurrency to 1 by setting --max-concurrency 1 in the CLI, or by setting evaluateOptions.maxConcurrency in the config.
2. Set a delay between requests by setting --delay 3000 (3000 ms) in the CLI, or by setting evaluateOptions.delay in the config, or with the environment variable PROMPTFOO_DELAY_MS (all values are in milliseconds).
3. Adjust the exponential backoff for failed requests by setting the environment variable PROMPTFOO_REQUEST_BACKOFF_MS. This defaults to 5000 milliseconds and retries exponential up to 4 times. You can increase this value if requests are still failing, but note that this can significantly increase end-to-end test time.

OpenAI flakiness

To retry HTTP requests that are Internal Server errors, set the PROMPTFOO_RETRY_5XX environment variable to 1.