Quickstart

Connect through a coding harness, an API or SDK, or an LLM gateway. Harnesses and direct API clients use the same Fireworks model IDs and account controls. Gateway setup varies by provider.

Coding harness
API or SDK
LLM gateway

Install FireConnect, sign in, and connect the harness you already use:

curl -fsSL \
  https://raw.githubusercontent.com/fw-ai/fireconnect/main/install.sh \
  | bash

fireconnect login
fireconnect claude

Restart Claude Code, then open /model. FireConnect automatically adds coding-ready open models and auto. Replace claude with another harness from Coding Harnesses.

Point an OpenAI-compatible client at Fireworks and choose a model:

pip install openai
export FIREWORKS_API_KEY="YOUR_FIREWORKS_API_KEY"

import os

from openai import OpenAI

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

response = client.responses.create(
    model="glm-latest",
    input="Review this pull request.",
)

print(response.output_text)

Fireworks also supports Chat Completions and Anthropic Messages. See APIs and SDKs.

Keep your existing gateway and configure its Fireworks provider. For LiteLLM, create config.yaml:

model_list:
  - model_name: glm-latest
    litellm_params:
      model: fireworks_ai/glm-latest
      api_key: os.environ/FIREWORKS_API_KEY
      api_base: https://api.fireworks.ai/inference/v1

Start the proxy:

pip install "litellm[proxy]"
export FIREWORKS_API_KEY="YOUR_FIREWORKS_API_KEY"
litellm --config config.yaml

In another terminal, send a request:

curl http://localhost:4000/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-latest",
    "messages": [{"role": "user", "content": "Say pong in one word."}]
  }'

Provider names, key configuration, and model catalogs vary by gateway. See LLM Gateways.

Choose models or manage usage and spend

Open Models

Use auto, follow a model family, choose a fast tier, or pin a version.

FireRouter

Let one model ID choose a model for each new user turn.

Usage and Cost

See session estimates and Fireworks model usage by model, user, or API key.

Spend Limits

Set per-user caps for supported Fireworks serverless models, with account defaults, group limits, and user exceptions.

Nexus

Connect

Choose models

Operate

Reference

Choose models or manage usage and spend

Open Models

FireRouter

Usage and Cost

Spend Limits

​Choose models or manage usage and spend

Open Models

FireRouter

Usage and Cost

Spend Limits

Choose models or manage usage and spend