Fireworks AI -- Fast Inference & Fine-Tuning

# Verify credentials
[ -n "$FIREWORKS_API_KEY" ] && echo "FIREWORKS_API_KEY set" || echo "NOT SET"

pip install fireworks-ai openai

import os
os.environ["FIREWORKS_API_KEY"] = "fw_..."  # from https://fireworks.ai/api-keys

from openai import OpenAI

client = OpenAI(
    base_url="https://api.fireworks.ai/inference/v1",
    api_key=os.environ["FIREWORKS_API_KEY"],
)

response = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain gradient descent in two sentences."},
    ],
    max_tokens=256,
    temperature=0.7,
)
print(response.choices[0].message.content)

from openai import OpenAI

client = OpenAI(
    base_url="https://api.fireworks.ai/inference/v1",
    api_key=os.environ["FIREWORKS_API_KEY"],
)

response = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[
        {"role": "user", "content": "Write a Python quicksort function."},
    ],
    max_tokens=512,
    temperature=0.0,
)
print(response.choices[0].message.content)

stream = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[{"role": "user", "content": "Explain transformers."}],
    stream=True,
    max_tokens=512,
)
for chunk in stream:
    delta = chunk.choices[0].delta.content
    if delta:
        print(delta, end="", flush=True)

tools = [
    {
        "type": "function",
        "function": {
            "name": "get_weather",
            "description": "Get current weather for a location",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {"type": "string", "description": "City name"},
                    "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
                },
                "required": ["location"],
            },
        },
    }
]

response = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[{"role": "user", "content": "What is the weather in San Francisco?"}],
    tools=tools,
    tool_choice="auto",
)
tool_call = response.choices[0].message.tool_calls[0]
print(tool_call.function.name, tool_call.function.arguments)

response = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[{"role": "user", "content": "List 3 planets with mass and diameter."}],
    response_format={"type": "json_object"},
    max_tokens=512,
)
import json
data = json.loads(response.choices[0].message.content)

response = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[{"role": "user", "content": "Extract name and age from: John is 30."}],
    response_format={
        "type": "json_object",
        "schema": {
            "type": "object",
            "properties": {
                "name": {"type": "string"},
                "age": {"type": "integer"},
            },
            "required": ["name", "age"],
        },
    },
)

response = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p2-11b-vision-instruct",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "Describe this image."},
                {
                    "type": "image_url",
                    "image_url": {"url": "https://example.com/photo.jpg"},
                },
            ],
        }
    ],
    max_tokens=256,
)
print(response.choices[0].message.content)

{"messages": [{"role": "system", "content": "You are a coding assistant."}, {"role": "user", "content": "Write a Python hello world."}, {"role": "assistant", "content": "print('Hello, world!')"}]}
{"messages": [{"role": "user", "content": "What is 2+2?"}, {"role": "assistant", "content": "4"}]}

import requests

url = "https://api.fireworks.ai/v1/accounts/{account_id}/supervisedFineTuningJobs"
headers = {
    "Authorization": f"Bearer {os.environ['FIREWORKS_API_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "displayName": "my-sft-job",
    "model": "accounts/fireworks/models/llama-v3p3-70b-instruct",
    "dataset": "accounts/{account_id}/datasets/{dataset_id}",
    "hyperparameters": {
        "epochs": 3,
        "learning_rate": 1e-4,
        "batch_size": 8,
        "lora_rank": 16,
    },
}
response = requests.post(url, headers=headers, json=payload)
job = response.json()
print(f"Job ID: {job['name']}")

job_url = f"https://api.fireworks.ai/v1/{job['name']}"
status = requests.get(job_url, headers=headers).json()
print(f"State: {status['state']}, Progress: {status.get('progress', 'N/A')}")

{"chosen": [{"role": "user", "content": "Explain ML"}, {"role": "assistant", "content": "Machine learning is..."}], "rejected": [{"role": "user", "content": "Explain ML"}, {"role": "assistant", "content": "ML is complicated..."}]}

url = "https://api.fireworks.ai/v1/accounts/{account_id}/dpoFineTuningJobs"
payload = {
    "displayName": "my-dpo-job",
    "model": "accounts/fireworks/models/llama-v3p3-70b-instruct",
    "dataset": "accounts/{account_id}/datasets/{dataset_id}",
    "hyperparameters": {
        "epochs": 2,
        "learning_rate": 5e-5,
        "beta": 0.1,
    },
}
response = requests.post(url, headers=headers, json=payload)

url = "https://api.fireworks.ai/v1/accounts/{account_id}/reinforcementFineTuningJobs"
payload = {
    "displayName": "my-rl-job",
    "baseModel": "accounts/fireworks/models/llama-v3p3-70b-instruct",
    "rewardModel": "accounts/{account_id}/models/{reward_model_id}",
    "dataset": "accounts/{account_id}/datasets/{dataset_id}",
}
response = requests.post(url, headers=headers, json=payload)

import requests

url = "https://api.fireworks.ai/v1/accounts/{account_id}/deployments"
headers = {
    "Authorization": f"Bearer {os.environ['FIREWORKS_API_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "displayName": "my-llama-deployment",
    "model": "accounts/fireworks/models/llama-v3p3-70b-instruct",
    "deploymentShape": "fast",  # Options: fast, throughput, minimal
    "minReplicaCount": 1,
    "maxReplicaCount": 4,
}
response = requests.post(url, headers=headers, json=payload)
deployment = response.json()

# List deployments
deployments = requests.get(
    f"https://api.fireworks.ai/v1/accounts/{{account_id}}/deployments",
    headers=headers,
).json()

# Scale deployment
requests.patch(
    f"https://api.fireworks.ai/v1/{deployment['name']}",
    headers=headers,
    json={"minReplicaCount": 2, "maxReplicaCount": 8},
)

# Delete deployment
requests.delete(
    f"https://api.fireworks.ai/v1/{deployment['name']}",
    headers=headers,
)

response = client.chat.completions.create(
    model="accounts/{account_id}/deployments/{deployment_id}",
    messages=[{"role": "user", "content": "Hello!"}],
)

response = client.embeddings.create(
    model="nomic-ai/nomic-embed-text-v1.5",
    input=["Machine learning is a subset of AI.", "Deep learning uses neural networks."],
)
for i, emb in enumerate(response.data):
    print(f"Embedding {i}: {len(emb.embedding)} dimensions")

# macOS / Linux (Homebrew)
brew tap fw-ai/firectl && brew install firectl

# Install script
curl -sSL https://cli.fireworks.ai/install.sh | bash

# Windows (Chocolatey)
choco install firectl

# Verify
firectl version

# Upgrade
firectl upgrade

firectl signin      # Interactive login
firectl whoami      # Show current account

# Upload a custom model
firectl model create my-model /path/to/model/weights

# List models
firectl model list

# Get model details
firectl model get accounts/{account_id}/models/my-model

# Delete model
firectl model delete accounts/{account_id}/models/my-model

# Create on-demand deployment
firectl deployment create accounts/fireworks/models/llama-v3p3-70b-instruct \
    --display-name "prod-llama"

# List deployments
firectl deployment list

# Scale deployment
firectl deployment scale {deployment_id} \
    --min-replica-count 2 --max-replica-count 8

# Delete deployment
firectl deployment delete {deployment_id}

# Create SFT job
firectl supervised-fine-tuning-job create my-sft-job \
    --model accounts/fireworks/models/llama-v3p3-70b-instruct \
    --dataset accounts/{account_id}/datasets/my-dataset

# Create RL fine-tuning job
firectl reinforcement-fine-tuning-job create my-rl-job \
    --base-model accounts/fireworks/models/llama-v3p3-70b-instruct \
    --reward-model accounts/{account_id}/models/my-reward-model

# Monitor jobs
firectl fine-tuning-job list
firectl fine-tuning-job get my-sft-job

# Stop / resume
firectl fine-tuning-job stop my-sft-job
firectl fine-tuning-job resume my-sft-job

from openai import OpenAI

client = OpenAI(
    base_url="https://api.fireworks.ai/inference/v1",
    api_key=os.environ["FIREWORKS_API_KEY"],
)

# Chat completions -- same API as OpenAI
response = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[{"role": "user", "content": "Hello!"}],
)

# Streaming -- same API
stream = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True,
)

# Embeddings -- same API
embeddings = client.embeddings.create(
    model="nomic-ai/nomic-embed-text-v1.5",
    input=["text to embed"],
)

export OPENAI_API_BASE="https://api.fireworks.ai/inference/v1"
export OPENAI_API_KEY="fw_..."

response = client.chat.completions.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[{"role": "user", "content": "..."}],
    max_tokens=512,
    extra_body={"context_length_exceeded_behavior": "truncate"},  # or "error"
)

import fireworks.client

fireworks.client.api_key = os.environ["FIREWORKS_API_KEY"]

response = fireworks.client.ChatCompletion.create(
    model="accounts/fireworks/models/llama-v3p3-70b-instruct",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

# Prepare batch file (JSONL)
# Each line: {"custom_id": "req-1", "method": "POST", "url": "/v1/chat/completions", "body": {...}}

# Upload batch file
batch_file = client.files.create(
    file=open("batch_requests.jsonl", "rb"),
    purpose="batch",
)

# Create batch job
batch = client.batches.create(
    input_file_id=batch_file.id,
    endpoint="/v1/chat/completions",
    completion_window="24h",
)
print(f"Batch ID: {batch.id}, Status: {batch.status}")

# Check status
batch_status = client.batches.retrieve(batch.id)
print(f"Status: {batch_status.status}")