### Clone and Setup Repository

Source: https://docs.fireworks.ai/fine-tuning/quickstart-math

Commands to clone the GSM8K quickstart repository, navigate into the directory, and install the required Python dependencies from the requirements file.

```shellscript
git clone https://github.com/eval-protocol/quickstart-gsm8k.git
cd quickstart-gsm8k
pip install -r requirements.txt
```

--------------------------------

### Setup Eval Protocol Environment via CLI

Source: https://docs.fireworks.ai/llms-full.txt

Commands to clone the quickstart repository, install necessary dependencies, and initialize the required directory structure for GSM8K evaluation.

```bash
git clone https://github.com/eval-protocol/quickstart-gsm8k.git
cd quickstart-gsm8k
pip install -r requirements.txt
mkdir -p gsm8k_artifacts/{tests/pytest/gsm8k,development}
cp evaluation.py gsm8k_artifacts/tests/pytest/gsm8k/test_pytest_math_example.py
cp gsm8k_sample.jsonl gsm8k_artifacts/development/gsm8k_sample.jsonl
```

--------------------------------

### Install and Configure Eval Protocol

Source: https://docs.fireworks.ai/llms-full.txt

Commands to clone the quickstart repository, install the necessary Python package, and set up the environment configuration file for API keys.

```bash
git clone git@github.com:eval-protocol/quickstart.git
cd quickstart
pip install "eval-protocol[svgbench]"
cp evaluator/env.example evaluator/.env
```

--------------------------------

### GET /v1/accounts/{account_id}/supervisedFineTuningJobs

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Retrieves a list of all supervised fine-tuning jobs associated with the account.

```APIDOC
## GET /v1/accounts/{account_id}/supervisedFineTuningJobs

### Description
Lists all supervised fine-tuning jobs for the given account.

### Method
GET

### Endpoint
/v1/accounts/{account_id}/supervisedFineTuningJobs

### Parameters
#### Path Parameters
- **account_id** (string) - Required - The unique identifier of the account.

### Response
#### Success Response (200)
- **jobs** (array) - A list of fine-tuning job objects.
```

--------------------------------

### Fine-tuning Cookbook Examples

Source: https://docs.fireworks.ai/deployments/benchmarking

This section provides practical examples and guides for various fine-tuning techniques using the Fireworks AI training SDK. It covers general cookbook overviews, Reinforcement Learning (RL), Direct Preference Optimization (DPO), and Supervised Fine-Tuning (SFT) methodologies.

```json
{
  "source": "/fine-tuning/training-sdk/cookbook-overview",
  "destination": "/fine-tuning/training-sdk/cookbook/overview"
}
```

```json
{
  "source": "/fine-tuning/training-sdk/cookbook-rl",
  "destination": "/fine-tuning/training-sdk/cookbook/rl"
}
```

```json
{
  "source": "/fine-tuning/training-sdk/cookbook-dpo",
  "destination": "/fine-tuning/training-sdk/cookbook/dpo"
}
```

```json
{
  "source": "/fine-tuning/training-sdk/cookbook-sft",
  "destination": "/fine-tuning/training-sdk/cookbook/sft"
}
```

```json
{
  "source": "/fine-tuning/training-sdk/cookbook-reference",
  "destination": "/fine-tuning/training-sdk/cookbook/reference"
}
```

--------------------------------

### Setup and API Request Logging in Python

Source: https://docs.fireworks.ai/api-reference/audio-translations

Demonstrates the installation of required Python packages and the implementation of a logging mechanism to track API response times and text output. This snippet uses the OpenAI SDK to interface with LLM endpoints.

```python
!pip install openai requests

from openai import OpenAI
import requests
from dotenv import import

print(f"Took: {time.time() - start:.3f}s. Text: '{r.text}'")
```

--------------------------------

### Install Fireworks AI Python SDK (Alpha)

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Installs the alpha version of the Fireworks AI Python SDK. Requires Python 3.9+ and an API key. Use the `--pre` flag for the latest pre-release versions.

```bash
pip install --pre fireworks-ai
```

--------------------------------

### Reservation Commands

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Commands for retrieving information about reservations.

```APIDOC
## firectl reservation get

### Description
Prints information about a reservation.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl reservation list

### Description
Prints active reservations.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified
```

--------------------------------

### Deployments API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage and deploy models.

```APIDOC
## POST /v1/deployments

### Description
Creates a new deployment.

### Method
POST

### Endpoint
/v1/deployments

### Parameters
#### Request Body
- **name** (string) - Required - The name of the deployment.
- **model_id** (string) - Required - The ID of the model to deploy.
- **scale_type** (string) - Optional - The type of scaling.

### Request Example
```json
{
  "name": "my-llama-deployment",
  "model_id": "llama-v2-7b",
  "scale_type": "auto"
}
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the deployment.
- **name** (string) - The name of the deployment.
- **status** (string) - The status of the deployment.

#### Response Example
```json
{
  "id": "dep_abc123",
  "name": "my-llama-deployment",
  "status": "creating"
}
```
```

```APIDOC
## GET /v1/deployments

### Description
Lists all deployments.

### Method
GET

### Endpoint
/v1/deployments

### Response
#### Success Response (200)
- **deployments** (array) - A list of deployments.

#### Response Example
```json
{
  "deployments": [
    {
      "id": "dep_abc123",
      "name": "my-llama-deployment",
      "status": "running"
    }
  ]
}
```
```

```APIDOC
## GET /v1/deployments/{deployment_id}

### Description
Retrieves a specific deployment by its ID.

### Method
GET

### Endpoint
/v1/deployments/{deployment_id}

### Parameters
#### Path Parameters
- **deployment_id** (string) - Required - The ID of the deployment to retrieve.

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the deployment.
- **name** (string) - The name of the deployment.
- **status** (string) - The status of the deployment.

#### Response Example
```json
{
  "id": "dep_abc123",
  "name": "my-llama-deployment",
  "status": "running"
}
```
```

```APIDOC
## PUT /v1/deployments/{deployment_id}

### Description
Updates a specific deployment.

### Method
PUT

### Endpoint
/v1/deployments/{deployment_id}

### Parameters
#### Path Parameters
- **deployment_id** (string) - Required - The ID of the deployment to update.
#### Request Body
- **scale_type** (string) - Optional - The new scale type.

### Request Example
```json
{
  "scale_type": "fixed"
}
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the deployment.
- **name** (string) - The name of the deployment.
- **status** (string) - The status of the deployment.

#### Response Example
```json
{
  "id": "dep_abc123",
  "name": "my-llama-deployment",
  "status": "updating"
}
```
```

```APIDOC
## DELETE /v1/deployments/{deployment_id}

### Description
Deletes a specific deployment by its ID.

### Method
DELETE

### Endpoint
/v1/deployments/{deployment_id}

### Parameters
#### Path Parameters
- **deployment_id** (string) - Required - The ID of the deployment to delete.

### Response
#### Success Response (200)
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "message": "Deployment deleted successfully."
}
```
```

```APIDOC
## POST /v1/deployments/{deployment_id}/undelete

### Description
Undeletes a specific deployment.

### Method
POST

### Endpoint
/v1/deployments/{deployment_id}/undelete

### Parameters
#### Path Parameters
- **deployment_id** (string) - Required - The ID of the deployment to undelete.

### Response
#### Success Response (200)
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "message": "Deployment undeleted successfully."
}
```
```

```APIDOC
## POST /v1/deployments/{deployment_id}/scale

### Description
Scales a specific deployment.

### Method
POST

### Endpoint
/v1/deployments/{deployment_id}/scale

### Parameters
#### Path Parameters
- **deployment_id** (string) - Required - The ID of the deployment to scale.
#### Request Body
- **scale_type** (string) - Required - The new scale type.
- **min_replicas** (integer) - Optional - Minimum number of replicas.
- **max_replicas** (integer) - Optional - Maximum number of replicas.

### Request Example
```json
{
  "scale_type": "fixed",
  "min_replicas": 1,
  "max_replicas": 5
}
```

### Response
#### Success Response (200)
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "message": "Deployment scaling initiated."
}
```
```

--------------------------------

### Audio Batch API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage batch requests for audio processing.

```APIDOC
## Audio Batch API Endpoints

### Description
This section details the API endpoints for managing batch requests related to audio processing.

### Endpoints

- **POST /api-reference/create-batch-request**
  - **Description**: Creates a new batch request for audio processing.
- **GET /api-reference/get-batch-status**
  - **Description**: Retrieves the status of a batch request.
```

--------------------------------

### Models API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage and interact with models.

```APIDOC
## POST /v1/models

### Description
Creates a new model.

### Method
POST

### Endpoint
/v1/models

### Parameters
#### Request Body
- **name** (string) - Required - The name of the model.
- **source_url** (string) - Required - The URL of the model source.

### Request Example
```json
{
  "name": "my-custom-model",
  "source_url": "s3://my-bucket/my-model.gguf"
}
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the model.
- **name** (string) - The name of the model.
- **status** (string) - The status of the model.

#### Response Example
```json
{
  "id": "model_xyz789",
  "name": "my-custom-model",
  "status": "uploading"
}
```
```

```APIDOC
## GET /v1/models

### Description
Lists all models.

### Method
GET

### Endpoint
/v1/models

### Response
#### Success Response (200)
- **models** (array) - A list of models.

#### Response Example
```json
{
  "models": [
    {
      "id": "model_xyz789",
      "name": "my-custom-model",
      "status": "ready"
    }
  ]
}
```
```

```APIDOC
## GET /v1/models/{model_id}

### Description
Retrieves a specific model by its ID.

### Method
GET

### Endpoint
/v1/models/{model_id}

### Parameters
#### Path Parameters
- **model_id** (string) - Required - The ID of the model to retrieve.

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the model.
- **name** (string) - The name of the model.
- **status** (string) - The status of the model.

#### Response Example
```json
{
  "id": "model_xyz789",
  "name": "my-custom-model",
  "status": "ready"
}
```
```

```APIDOC
## GET /v1/models/{model_id}/upload

### Description
Gets the upload endpoint for a specific model.

### Method
GET

### Endpoint
/v1/models/{model_id}/upload

### Parameters
#### Path Parameters
- **model_id** (string) - Required - The ID of the model.

### Response
#### Success Response (200)
- **upload_url** (string) - The URL to upload the model files to.

#### Response Example
```json
{
  "upload_url": "https://s3.amazonaws.com/my-bucket/uploads/model_xyz789?signature=..."
}
```
```

```APIDOC
## GET /v1/models/{model_id}/download

### Description
Gets the download endpoint for a specific model.

### Method
GET

### Endpoint
/v1/models/{model_id}/download

### Parameters
#### Path Parameters
- **model_id** (string) - Required - The ID of the model.

### Response
#### Success Response (200)
- **download_url** (string) - The URL to download the model files from.

#### Response Example
```json
{
  "download_url": "https://s3.amazonaws.com/my-bucket/models/model_xyz789.gguf"
}
```
```

```APIDOC
## PUT /v1/models/{model_id}

### Description
Updates a specific model.

### Method
PUT

### Endpoint
/v1/models/{model_id}

### Parameters
#### Path Parameters
- **model_id** (string) - Required - The ID of the model to update.
#### Request Body
- **name** (string) - Optional - The new name for the model.

### Request Example
```json
{
  "name": "my-updated-model-name"
}
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the model.
- **name** (string) - The updated name of the model.
- **status** (string) - The status of the model.

#### Response Example
```json
{
  "id": "model_xyz789",
  "name": "my-updated-model-name",
  "status": "ready"
}
```
```

```APIDOC
## DELETE /v1/models/{model_id}

### Description
Deletes a specific model by its ID.

### Method
DELETE

### Endpoint
/v1/models/{model_id}

### Parameters
#### Path Parameters
- **model_id** (string) - Required - The ID of the model to delete.

### Response
#### Success Response (200)
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "message": "Model deleted successfully."
}
```
```

```APIDOC
## POST /v1/models/{model_id}/prepare

### Description
Prepares a model for use.

### Method
POST

### Endpoint
/v1/models/{model_id}/prepare

### Parameters
#### Path Parameters
- **model_id** (string) - Required - The ID of the model to prepare.

### Response
#### Success Response (200)
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "message": "Model preparation initiated."
}
```
```

```APIDOC
## POST /v1/models/validate/upload

### Description
Validates a model upload.

### Method
POST

### Endpoint
/v1/models/validate/upload

### Parameters
#### Request Body
- **model_id** (string) - Required - The ID of the model.
- **file_url** (string) - Required - The URL of the uploaded model file.

### Request Example
```json
{
  "model_id": "model_xyz789",
  "file_url": "s3://my-bucket/uploads/model_xyz789?signature=..."
}
```

### Response
#### Success Response (200)
- **message** (string) - Validation status message.

#### Response Example
```json
{
  "message": "Model upload validation successful."
}
```
```

--------------------------------

### POST /v1/chat/completions

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Create a chat completion for the provided prompt and parameters using the OpenAI-compatible API.

```APIDOC
## POST /v1/chat/completions

### Description
Create a completion for the provided prompt and parameters.

### Method
POST

### Endpoint
/v1/chat/completions

### Request Body
- **model** (string) - Required - The ID of the model to use.
- **messages** (array) - Required - A list of messages comprising the conversation so far.

### Request Example
{
  "model": "accounts/fireworks/models/llama-v3-70b-instruct",
  "messages": [{"role": "user", "content": "Hello!"}]
}

### Response
#### Success Response (200)
- **id** (string) - Unique identifier for the completion.
- **choices** (array) - List of completion choices.

#### Response Example
{
  "id": "chatcmpl-123",
  "choices": [{"message": {"role": "assistant", "content": "Hello there!"}}]
}
```

--------------------------------

### Accounts API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage user accounts within the system.

```APIDOC
## Accounts API Endpoints

### Description
This section provides endpoints for managing user accounts, including retrieval and listing of account information.

### Endpoints

- **GET /api-reference/accounts**
  - **Description**: Retrieves a list of user accounts.
```

--------------------------------

### POST /v1/accounts/{account_id}/supervisedFineTuningJobs

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Creates a new supervised fine-tuning job within the specified account.

```APIDOC
## POST /v1/accounts/{account_id}/supervisedFineTuningJobs

### Description
Initiates a new supervised fine-tuning training process.

### Method
POST

### Endpoint
/v1/accounts/{account_id}/supervisedFineTuningJobs

### Parameters
#### Path Parameters
- **account_id** (string) - Required - The unique identifier of the account.

### Response
#### Success Response (200)
- **job_id** (string) - The unique identifier of the created job.
```

--------------------------------

### Responses API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage and retrieve responses generated by the API.

```APIDOC
## POST /v1/responses

### Description
Creates a new response.

### Method
POST

### Endpoint
/v1/responses

### Parameters
#### Request Body
- **content** (string) - Required - The content of the response.

### Request Example
```json
{
  "content": "This is a sample response."
}
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the response.
- **content** (string) - The content of the response.

#### Response Example
```json
{
  "id": "resp_abc123",
  "content": "This is a sample response."
}
```
```

```APIDOC
## GET /v1/responses

### Description
Lists all responses.

### Method
GET

### Endpoint
/v1/responses

### Response
#### Success Response (200)
- **responses** (array) - A list of responses.

#### Response Example
```json
{
  "responses": [
    {
      "id": "resp_abc123",
      "content": "This is a sample response."
    }
  ]
}
```
```

```APIDOC
## GET /v1/responses/{response_id}

### Description
Retrieves a specific response by its ID.

### Method
GET

### Endpoint
/v1/responses/{response_id}

### Parameters
#### Path Parameters
- **response_id** (string) - Required - The ID of the response to retrieve.

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the response.
- **content** (string) - The content of the response.

#### Response Example
```json
{
  "id": "resp_abc123",
  "content": "This is a sample response."
}
```
```

```APIDOC
## DELETE /v1/responses/{response_id}

### Description
Deletes a specific response by its ID.

### Method
DELETE

### Endpoint
/v1/responses/{response_id}

### Parameters
#### Path Parameters
- **response_id** (string) - Required - The ID of the response to delete.

### Response
#### Success Response (200)
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "message": "Response deleted successfully."
}
```
```

--------------------------------

### Deployment Shapes API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Retrieve information about available deployment shapes and their versions.

```APIDOC
## Deployment Shapes API Endpoints

### Description
This section provides endpoints to query available deployment shapes and their specific versions.

### Endpoints

- **GET /api-reference/list-deployment-shape-versions**
  - **Description**: Lists all available versions for deployment shapes.
- **GET /api-reference/get-deployment-shape**
  - **Description**: Retrieves details of a specific deployment shape.
- **GET /api-reference/get-deployment-shape-version**
  - **Description**: Retrieves details of a specific version of a deployment shape.
```

--------------------------------

### Datasets API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage datasets for fine-tuning, including creation, upload, and retrieval.

```APIDOC
## Datasets API Endpoints

### Description
This section details the API endpoints for managing datasets used in fine-tuning processes, covering creation, file handling, and validation.

### Endpoints

- **POST /api-reference/create-dataset**
  - **Description**: Creates a new dataset.
- **GET /api-reference/list-datasets**
  - **Description**: Lists all available datasets.
- **GET /api-reference/get-dataset**
  - **Description**: Retrieves details of a specific dataset.
- **GET /api-reference/get-dataset-upload-endpoint**
  - **Description**: Gets the endpoint for uploading dataset files.
- **GET /api-reference/get-dataset-download-endpoint**
  - **Description**: Gets the endpoint for downloading dataset files.
- **PUT /api-reference/update-dataset**
  - **Description**: Updates an existing dataset.
- **DELETE /api-reference/delete-dataset**
  - **Description**: Deletes a specified dataset.
- **POST /api-reference/upload-dataset-files**
  - **Description**: Uploads files to a dataset.
- **POST /api-reference/validate-dataset-upload**
  - **Description**: Validates a dataset upload.
```

--------------------------------

### LoRA API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage LoRA (Low-Rank Adaptation) models.

```APIDOC
## POST /v1/lora/models

### Description
Creates a new LoRA model.

### Method
POST

### Endpoint
/v1/lora/models

### Parameters
#### Request Body
- **name** (string) - Required - The name of the LoRA model.
- **base_model_id** (string) - Required - The ID of the base model.
- **lora_model_url** (string) - Required - The URL of the LoRA model weights.

### Request Example
```json
{
  "name": "my-lora-model",
  "base_model_id": "llama-v2-7b",
  "lora_model_url": "s3://my-bucket/lora/my-lora.safetensors"
}
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the LoRA model.
- **name** (string) - The name of the LoRA model.

#### Response Example
```json
{
  "id": "lora_abc123",
  "name": "my-lora-model"
}
```
```

```APIDOC
## GET /v1/lora/models

### Description
Lists all LoRA models.

### Method
GET

### Endpoint
/v1/lora/models

### Response
#### Success Response (200)
- **lora_models** (array) - A list of LoRA models.

#### Response Example
```json
{
  "lora_models": [
    {
      "id": "lora_abc123",
      "name": "my-lora-model"
    }
  ]
}
```
```

```APIDOC
## GET /v1/lora/models/{lora_model_id}

### Description
Retrieves a specific LoRA model by its ID.

### Method
GET

### Endpoint
/v1/lora/models/{lora_model_id}

### Parameters
####
```

--------------------------------

### Reinforcement Fine-tuning Steps API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage individual steps within reinforcement fine-tuning processes.

```APIDOC
## Reinforcement Fine-tuning Steps API Endpoints

### Description
This section covers the API endpoints for managing and executing individual steps within reinforcement fine-tuning.

### Endpoints

- **POST /api-reference/create-reinforcement-fine-tuning-step**
  - **Description**: Creates a new step for reinforcement fine-tuning.
- **GET /api-reference/list-reinforcement-fine-tuning-steps**
  - **Description**: Lists all reinforcement fine-tuning steps.
- **GET /api-reference/get-reinforcement-fine-tuning-step**
  - **Description**: Retrieves details of a specific reinforcement fine-tuning step.
- **DELETE /api-reference/delete-reinforcement-fine-tuning-step**
  - **Description**: Deletes a specified reinforcement fine-tuning step.
- **POST /api-reference/resume-reinforcement-fine-tuning-step**
  - **Description**: Resumes a paused reinforcement fine-tuning step.
- **POST /api-reference/execute-reinforcement-fine-tuning-step**
  - **Description**: Executes a reinforcement fine-tuning step.
```

--------------------------------

### Audio API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Process audio data, including transcriptions and translations.

```APIDOC
## Audio API Endpoints

### Description
This section covers the API endpoints for various audio processing tasks, such as real-time transcriptions, batch transcriptions, and translations.

### Endpoints

- **POST /api-reference/audio-streaming-transcriptions**
  - **Description**: Performs streaming audio transcriptions.
- **POST /api-reference/audio-transcriptions**
  - **Description**: Performs batch audio transcriptions.
- **POST /api-reference/audio-translations**
  - **Description**: Translates audio content.
```

--------------------------------

### Quota Management Commands

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Commands for retrieving and updating quota information.

```APIDOC
## firectl quota get

### Description
Prints information about a quota.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl quota list

### Description
Prints all quotas.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl quota update

### Description
Updates a quota.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified
```

--------------------------------

### Models API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage models, including creation, retrieval, updates, and deletion.

```APIDOC
## Models API Endpoints

### Description
This section covers the API endpoints for managing machine learning models, including their lifecycle and associated endpoints.

### Endpoints

- **POST /api-reference/create-model**
  - **Description**: Creates a new model.
- **GET /api-reference/list-models**
  - **Description**: Lists all available models.
- **GET /api-reference/get-model**
  - **Description**: Retrieves details of a specific model.
- **GET /api-reference/get-model-upload-endpoint**
  - **Description**: Gets the endpoint for uploading model files.
- **GET /api-reference/get-model-download-endpoint**
  - **Description**: Gets the endpoint for downloading model files.
- **PUT /api-reference/update-model**
  - **Description**: Updates an existing model.
- **DELETE /api-reference/delete-model**
  - **Description**: Deletes a specified model.
- **POST /api-reference/prepare-model**
  - **Description**: Prepares a model for deployment or further processing.
- **POST /api-reference/validate-model-upload**
  - **Description**: Validates a model upload.
```

--------------------------------

### Image API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Generate and edit images using text prompts and context.

```APIDOC
## Image API Endpoints

### Description
This section covers the API endpoints for image generation and editing, utilizing text prompts and contextual information.

### Endpoints

- **POST /api-reference/generate-a-new-image-from-a-text-prompt**
  - **Description**: Generates a new image from a given text prompt.
- **POST /api-reference/generate-or-edit-image-using-flux-kontext**
  - **Description**: Generates or edits an image using Flux context.
- **GET /api-reference/get-generated-image-from-flux-kontex**
  - **Description**: Retrieves a generated image from Flux context.
```

--------------------------------

### Batch API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage and interact with batch inference jobs for large language models.

```APIDOC
## Batch API Endpoints

### Description
This section details the API endpoints for managing batch inference jobs, including creating, listing, retrieving, and deleting jobs.

### Endpoints

- **POST /api-reference/create-batch-inference-job**
  - **Description**: Creates a new batch inference job.
- **GET /api-reference/list-batch-inference-jobs**
  - **Description**: Lists all available batch inference jobs.
- **GET /api-reference/get-batch-inference-job**
  - **Description**: Retrieves details of a specific batch inference job.
- **DELETE /api-reference/delete-batch-inference-job**
  - **Description**: Deletes a specified batch inference job.
```

--------------------------------

### Deployments API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage and control model deployments, including creation, updates, and scaling.

```APIDOC
## Deployments API Endpoints

### Description
This section details the API endpoints for managing model deployments, including lifecycle management and scaling operations.

### Endpoints

- **POST /api-reference/create-deployment**
  - **Description**: Creates a new model deployment.
- **GET /api-reference/list-deployments**
  - **Description**: Lists all existing model deployments.
- **GET /api-reference/get-deployment**
  - **Description**: Retrieves details of a specific model deployment.
- **PUT /api-reference/update-deployment**
  - **Description**: Updates an existing model deployment.
- **DELETE /api-reference/delete-deployment**
  - **Description**: Deletes a specified model deployment.
- **POST /api-reference/undelete-deployment**
  - **Description**: Undeletes a previously deleted model deployment.
- **POST /api-reference/scale-deployment**
  - **Description**: Scales a model deployment up or down.
```

--------------------------------

### Model Management Commands

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Commands for loading, preparing, unloading, updating, and uploading models.

```APIDOC
## firectl model load-lora

### Description
Loads a LoRA model.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl model prepare

### Description
Prepare models for different precisions.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl model unload-lora

### Description
Unloads a LoRA model.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl model update

### Description
Updates a model.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl model upload

### Description
Resumes or completes a model upload.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified
```

--------------------------------

### Reinforcement Fine-tuning Jobs API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage reinforcement fine-tuning jobs, including creation, cancellation, and resumption.

```APIDOC
## Reinforcement Fine-tuning Jobs API Endpoints

### Description
This section details the API endpoints for managing reinforcement fine-tuning jobs, including their lifecycle and control operations.

### Endpoints

- **POST /api-reference/create-reinforcement-fine-tuning-job**
  - **Description**: Creates a new reinforcement fine-tuning job.
- **GET /api-reference/list-reinforcement-fine-tuning-jobs**
  - **Description**: Lists all reinforcement fine-tuning jobs.
- **GET /api-reference/get-reinforcement-fine-tuning-job**
  - **Description**: Retrieves details of a specific reinforcement fine-tuning job.
- **DELETE /api-reference/delete-reinforcement-fine-tuning-job**
  - **Description**: Deletes a specified reinforcement fine-tuning job.
- **POST /api-reference/cancel-reinforcement-fine-tuning-job**
  - **Description**: Cancels a running reinforcement fine-tuning job.
- **POST /api-reference/resume-reinforcement-fine-tuning-job**
  - **Description**: Resumes a paused reinforcement fine-tuning job.
```

--------------------------------

### POST /v1/chat/completions

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Creates a model response for the given chat conversation using an OpenAI-compatible interface.

```APIDOC
## POST /v1/chat/completions

### Description
Generates a chat completion using the specified model.

### Method
POST

### Endpoint
/v1/chat/completions

### Request Body
- **model** (string) - Required - The ID of the model to use.
- **messages** (array) - Required - A list of messages comprising the conversation.

### Request Example
{
  "model": "accounts/fireworks/models/llama-v3-70b-instruct",
  "messages": [{"role": "user", "content": "Hello!"}]
}

### Response
#### Success Response (200)
- **id** (string) - Unique identifier for the completion.
- **choices** (array) - List of generated completions.

#### Response Example
{
  "id": "chatcmpl-123",
  "choices": [{"message": {"role": "assistant", "content": "Hello! How can I help you?"}}]
}
```

--------------------------------

### POST /v1/messages

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Creates a message response using the Anthropic-compatible API format.

```APIDOC
## POST /v1/messages

### Description
Sends a message to the model using the Anthropic-compatible API structure.

### Method
POST

### Endpoint
/v1/messages

### Request Body
- **model** (string) - Required - The model to use.
- **messages** (array) - Required - Input messages.

### Request Example
{
  "model": "claude-3-opus",
  "messages": [{"role": "user", "content": "Explain quantum computing."}]
}

### Response
#### Success Response (200)
- **content** (array) - The generated response content.

#### Response Example
{
  "content": [{"type": "text", "text": "Quantum computing is..."}]
}
```

--------------------------------

### Bootstrap Training Environment

Source: https://docs.fireworks.ai/fine-tuning/training-sdk/training-and-sampling

This snippet demonstrates the initial setup phase for a training job, involving the initialization of the DeploymentManager and TrainerJobManager to prepare the environment for iterative updates.

```python
from fireworks.training import DeploymentManager, TrainerJobManager

# Initialize managers for the training lifecycle
deployment = DeploymentManager(shape_id="model-shape-v1")
trainer = TrainerJobManager(shape_id="model-shape-v1")

# Bootstrap resources
deployment.create()
trainer.create()
```

--------------------------------

### Set API Key Command

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Sets the default API key in the configuration file.

```APIDOC
## firectl set-api-key

### Description
Sets the default API key in ~/.fireworks/auth.ini.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified
```

--------------------------------

### Example Dockerfile for Fireworks AI

Source: https://docs.fireworks.ai/fine-tuning/quickstart-svg-agent

A sample Dockerfile demonstrating the setup of a Python 3.11-slim environment. It includes instructions for installing system dependencies like Chromium and managing Python packages via requirements.txt.

```docker
FROM python:3.11-slim

WORKDIR /app

# Install system dependencies
RUN apt-get update && apt-get install -y \
 chromium \
 && rm -rf /var/lib/apt/lists/*

# Install Python dependencies
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
```

--------------------------------

### Create a Datum from tokens and weights

Source: https://docs.fireworks.ai/fine-tuning/training-sdk/introduction

Demonstrates how to import the datum utility and prepare training data by tokenizing a prompt and response string. This snippet shows the initial setup required to create a Datum object for remote GPU training.

```python
import torch
from tinker_cookbook.supervised.common import datum_from_tokens_weights

tokens = tokenizer.encode("What is 2+2? The answer is 4.")
prompt_len = len(tokenizer.encode("What is 2+2?"))
```

--------------------------------

### Supervised Fine-tuning Jobs API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Manage supervised fine-tuning jobs, including creation, listing, and status updates.

```APIDOC
## Supervised Fine-tuning Jobs API Endpoints

### Description
This section covers the API endpoints for managing supervised fine-tuning jobs, from creation to deletion and resumption.

### Endpoints

- **POST /api-reference/create-supervised-fine-tuning-job**
  - **Description**: Creates a new supervised fine-tuning job.
- **GET /api-reference/list-supervised-fine-tuning-jobs**
  - **Description**: Lists all supervised fine-tuning jobs.
- **GET /api-reference/get-supervised-fine-tuning-job**
  - **Description**: Retrieves details of a specific supervised fine-tuning job.
- **DELETE /api-reference/delete-supervised-fine-tuning-job**
  - **Description**: Deletes a specified supervised fine-tuning job.
- **POST /api-reference/resume-supervised-fine-tuning-job**
  - **Description**: Resumes a paused supervised fine-tuning job.
```

--------------------------------

### OpenAI-compatible API

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Interact with models using an OpenAI-compatible API for chat completions and text generation.

```APIDOC
## POST /v1/chat/completions

### Description
Creates a chat completion choice for the provided prompt and parameters.

### Method
POST

### Endpoint
/v1/chat/completions

### Parameters
#### Request Body
- **model** (string) - Required - The model to use for completion.
- **messages** (array) - Required - A list of messages comprising the conversation so far.
- **temperature** (number) - Optional - Controls randomness. Lower values make the output more deterministic.
- **max_tokens** (integer) - Optional - The maximum number of tokens to generate.

### Request Example
```json
{
  "model": "accounts/fireworks/models/llama-v2-7b-chat",
  "messages": [
    {"role": "user", "content": "What is the capital of France?"}
  ],
  "max_tokens": 100,
  "temperature": 0.7
}
```

### Response
#### Success Response (200)
- **choices** (array) - A list of chat completion choices.
  - **message** (object) - The message content.
    - **role** (string) - The role of the author.
    - **content** (string) - The content of the message.

#### Response Example
```json
{
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "The capital of France is Paris."
      }
    }
  ]
}
```
```

```APIDOC
## POST /v1/completions

### Description
Creates a completion choice for the provided prompt and parameters.

### Method
POST

### Endpoint
/v1/completions

### Parameters
#### Request Body
- **model** (string) - Required - The model to use for completion.
- **prompt** (string) - Required - The prompt(s) to generate completions for.
- **temperature** (number) - Optional - Controls randomness. Lower values make the output more deterministic.
- **max_tokens** (integer) - Optional - The maximum number of tokens to generate.

### Request Example
```json
{
  "model": "accounts/fireworks/models/llama-v2-7b",
  "prompt": "Once upon a time",
  "max_tokens": 50,
  "temperature": 0.8
}
```

### Response
#### Success Response (200)
- **choices** (array) - A list of completion choices.
  - **text** (string) - The generated completion.

#### Response Example
```json
{
  "choices": [
    {
      "text": " there was a brave knight who lived in a faraway kingdom."
    }
  ]
}
```
```

--------------------------------

### Secret Management Commands

Source: https://docs.fireworks.ai/tools-sdks/python-sdk

Commands for creating, deleting, retrieving, listing, and updating secrets.

```APIDOC
## firectl secret create

### Description
Creates a secret for the signed in user.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl secret delete

### Description
Deletes a secret.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl secret get

### Description
Retrieves a secret by name.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl secret list

### Description
Lists all secrets for the signed in user.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified

---

## firectl secret update

### Description
Updates an existing secret.

### Method
Not specified (CLI command)

### Endpoint
Not applicable (CLI command)

### Parameters
Not specified

### Request Example
Not applicable

### Response
Not specified
```

--------------------------------

### Configure Model Deployment and Training

Source: https://docs.fireworks.ai/fine-tuning/training-sdk/training-and-sampling

This snippet illustrates setting up model deployment configurations, including replica counts, waiting for service readiness, and initializing a training job with a specific base model.

```python
base_model = "accounts/fireworks/models/qwen3-8b",
min_replica_count = 0,
max_replica_count = 1

deploy_mgr.wait_for_ready("research-serving")

# This is the only shape-specific value you choose
profile = rlor_mgr.resolve_training_profile(shape_id)

# Create trainer (polls until healthy)
endpoint = rlor_mgr.create_and_wait(TrainerJobConfig(
base_model = "accounts/fireworks/models/qwen3-8b"
```