Skip to main content
FireConnect is an open-source CLI that routes agentic coding harnesses through Fireworks models. Choose where inference runs:
  • Direct Fireworks routing (default) — requests go to the Fireworks gateway; use a Fireworks API key (fw_...) or Fire Pass key (fpk_...)
  • Fireworks on Microsoft Foundry — the same models, deployed in your Azure subscription and billed through Azure; use --provider azure and an Azure API key (see Fireworks on Microsoft Foundry)
Install once, configure your provider, then enable or disable routing per harness without editing config files by hand.

Claude Code

Anthropic-compatible routing with multi-slot model aliases

OpenCode

OpenAI-compatible adapter in opencode.json

Codex

OpenAI Codex CLI via the Responses API

Pi

Pi agent settings and auth

Cursor

OpenAI BYOK settings for Cursor IDE

VS Code

GitHub Copilot Chat custom endpoint

​
Prerequisites

​
Install

curl -fsSL https://raw.githubusercontent.com/fw-ai/fireconnect/main/install.sh | bash
For non-interactive setup: 
curl -fsSL https://raw.githubusercontent.com/fw-ai/fireconnect/main/install.sh | FIREWORKS_API_KEY="fw_..." bash
Fire Pass users can pass a fpk_... key directly. FireConnect detects the key type and applies the correct defaults. The installer:
  • Clones the CLI to ~/.fireconnect/cli and adds a fireconnect launcher to ~/.local/bin
  • Runs fireconnect configure to register harnesses and store your API key preference
  • Uses Node.js to update harness settings (it does not install or update npm packages)
After install, configure your provider (Fireworks is the default), then enable a harness: 
fireconnect configure --provider fireworks --api-key fw_...
fireconnect claude on      # Claude Code (default starting point)
fireconnect opencode on    # OpenCode
fireconnect codex on       # Codex
fireconnect pi on          # Pi
fireconnect cursor on      # Cursor IDE (quit Cursor first)
fireconnect vscode on      # VS Code + GitHub Copilot Chat

​
Upgrade

If you installed via the curl installer, pull the latest release in place: 
fireconnect upgrade
Re-run the install script if FireConnect was not installed with git (for example, a manual checkout). When a newer version is available, FireConnect prints a one-line update banner on startup with upgrade instructions. The check runs in the background and does not block your command. Check your installed version: 
fireconnect --version

​
Fireworks on Microsoft Foundry

Route OpenCode, Codex, and Pi through a Microsoft Foundry deployment of Fireworks models instead of the Fireworks gateway. Usage is billed through your Azure account and counts toward Azure commitments where applicable.

Microsoft Foundry integration guide

Portal setup, deployment modes (PayGo / PTU), billing, and troubleshooting — start here if you have not deployed a model yet
Terminology: In the CLI, the Foundry provider is --provider azure (or on --azure). Harness configs display the label Fireworks on Microsoft Foundry. Foundry uses deployment names you chose at deploy time (for example, FW-GLM-5.1) — not Fireworks serverless short IDs like glm-latest.
Claude Code, Cursor, and VS Code do not support Fireworks on Microsoft Foundry. Foundry exposes an OpenAI-compatible chat API, not the Anthropic Messages API Claude Code expects, and IDE BYOK flows require direct Fireworks credentials.

​
Prerequisites

  • A Microsoft Foundry resource with at least one Fireworks model deployment (for example, FW-GLM-5.1 or FW-MiniMax-M2.5)
  • Your Foundry resource endpoint and Azure API key from the Microsoft Foundry portal
  • OpenCode, Codex, or Pi installed locally
Follow the Microsoft Foundry integration guide and Microsoft Learn setup steps to enable Fireworks on Foundry and create a deployment before running FireConnect.

​
Configure the Foundry provider

Set the active provider once with configure. Harness on commands then use that provider — there is no separate global azure on command. 
export AZURE_API_KEY=<your-azure-api-key>

fireconnect configure \
  --provider azure \
  --base-url https://<resource>.services.ai.azure.com \
  --api-key $AZURE_API_KEY \
  --harnesses opencode,codex,pi
FireConnect normalizes --base-url to the OpenAI-compatible surface at .../openai/v1. You can paste the resource root, a project endpoint (.../api/projects/<name>), or an already-correct .../openai/v1 URL. To keep secrets out of ~/.fireconnect/config.json, export AZURE_API_KEY before configure instead of passing --api-key — FireConnect stores an environment reference.

​
Enable harnesses

After configuring the Foundry provider, enable harnesses the same way as with direct Fireworks routing: 
fireconnect opencode on
fireconnect codex on
fireconnect pi on
Each harness reads the provider, endpoint, and API key from ~/.fireconnect/config.json and writes OpenAI-compatible adapter settings. In harness UIs, the provider appears as Fireworks on Microsoft Foundry.

​
Choose a deployment

Pass the deployment name you chose in the Azure portal with --main: 
fireconnect opencode on --main FW-MiniMax-M2.5
fireconnect codex on --main FW-GLM-5.1
fireconnect pi on --main FW-GLM-5.1
If you omit --main, FireConnect defaults to FW-GLM-5.1.
fireconnect <harness> model list and model select browse the Fireworks serverless catalog (direct routing only). They do not list Foundry deployments. With --provider azure, set your deployment with --main on on.

​
One-off Foundry routing

To route a single harness through Foundry without changing global config, pass --azure on on: 
fireconnect opencode on \
  --azure \
  --base-url https://<resource>.services.ai.azure.com \
  --api-key $AZURE_API_KEY \
  --main FW-MiniMax-M2.5
If global config already has a Foundry endpoint, --azure alone reuses it: 
fireconnect opencode on --azure --main FW-GLM-5.1

​
Switch between direct Fireworks and Foundry

Update global config, then re-enable each harness: 
# Foundry
fireconnect configure \
  --provider azure \
  --base-url https://<resource>.services.ai.azure.com \
  --api-key $AZURE_API_KEY \
  --harnesses opencode,codex,pi

fireconnect opencode on --main FW-GLM-5.1
fireconnect codex on --main FW-GLM-5.1
fireconnect pi on --main FW-GLM-5.1

# Direct Fireworks
fireconnect configure --provider fireworks --api-key fw_...
fireconnect opencode on
fireconnect codex on
fireconnect pi on
Running off restores the harness config snapshot from before FireConnect was enabled for that provider.

​
API key resolution (Foundry)

When the Foundry provider is active, FireConnect resolves credentials in this order:
  1. Explicit --api-key on the command
  2. Global ~/.fireconnect/config.json
  3. AZURE_API_KEY environment variable

​
Verify routing

fireconnect opencode status   # provider=azure, base URL, deployment name
fireconnect codex status
fireconnect pi status
Turn off routing per harness with fireconnect <harness> off. This restores your previous config from the snapshot in ~/.fireconnect/<harness>/.

​
CLI design

FireConnect uses harness-first syntax: fireconnect <harness> <command> Bare harness names run on (for example, fireconnect claude is the same as fireconnect claude on).

​
Global commands

fireconnect configure    # Register harnesses, provider, and API key preferences
fireconnect upgrade      # Pull the latest FireConnect from GitHub (curl install)
fireconnect uninstall    # Disable all harnesses, restore configs, remove CLI
fireconnect help         # Show help
fireconnect --version    # Print the installed CLI version (-V also works)

​
Providers

FireConnect supports two inference backends. The CLI provider flag is shown in parentheses.
Provider flagWhere inference runsAPI keySupported harnesses
fireworks (default)Fireworks gatewayfw_... or fpk_...Claude Code, OpenCode, Codex, Pi, Cursor, VS Code
azureFireworks on Microsoft FoundryAzure API keyOpenCode, Codex, Pi
Set the default with fireconnect configure --provider fireworks or --provider azure. For Foundry, also pass --base-url with your resource endpoint. Harness on commands use the configured provider unless you pass --azure or per-command --base-url / --api-key overrides.

​
Per-harness commands

Each CLI harness (claude, opencode, codex, pi) supports:
  • fireconnect <harness> on — Route through the configured provider
  • fireconnect <harness> off — Restore your previous config
  • fireconnect <harness> status — Show provider, auth, and models
  • fireconnect <harness> model list — Browse Fireworks serverless models (direct Fireworks routing only)
  • fireconnect <harness> model select — Pick a serverless model interactively (direct Fireworks routing only)
  • fireconnect <harness> model reset — Reset models to defaults
  • fireconnect <harness> help — Harness-specific help
Each IDE harness (cursor, vscode) supports:
  • fireconnect <harness> on — Configure direct Fireworks routing (not Foundry; quit the IDE first)
  • fireconnect <harness> off — Restore your previous config (quit the IDE first)
  • fireconnect <harness> status — Show provider, auth, and registered models (read-only; IDE can stay open)
  • fireconnect <harness> model list — Browse Fireworks serverless models (read-only)
  • fireconnect <harness> model add <id> — Add a Fireworks model to the IDE picker
  • fireconnect <harness> model select — Pick a model interactively (Cursor: pass --mode)
  • fireconnect <harness> model reset — Reset fireconnect-managed models to defaults
  • fireconnect <harness> help — Harness-specific help
Cursor stores settings in SQLite (state.vscdb). Commands that write to that database (on, off, model add, model select, model reset) require Cursor to be fully quit first. VS Code stores the API key in state.vscdb as well, so on and off require quitting VS Code; model add, model select, and model reset only edit chatLanguageModels.json, which VS Code hot-reloads. With Fireworks on Microsoft Foundry (--provider azure), pass your Foundry deployment name to on with --main instead of using model list or model select. Run fireconnect help for the overview, or fireconnect claude help (and similarly for other harnesses) for harness-level options.

​
API key resolution

When a command needs credentials, FireConnect resolves them based on the active provider. Direct Fireworks routing (--provider fireworks)
  1. Explicit --api-key
  2. Harness-local stored key (for example, in ~/.claude/settings.json)
  3. Global ~/.fireconnect/config.json
  4. FIREWORKS_API_KEY environment variable
Fireworks on Microsoft Foundry (--provider azure)
  1. Explicit --api-key
  2. Global ~/.fireconnect/config.json
  3. AZURE_API_KEY environment variable
These serverless model short IDs apply to direct Fireworks routing and expand to full Fireworks paths automatically. With Fireworks on Microsoft Foundry, pass your deployment name instead (for example, FW-GLM-5.1 or FW-MiniMax-M2.5) via --main.
Short IDBest forNotes
glm-latestAll-around use, agentic tasksDefault for main and opus slots in Claude Code. Strong reasoning, 1M context.
glm-5p1General use (lighter)Default sonnet slot in Claude Code. Good balance of speed and quality.
deepseek-v4-flashBackground / fast tasksDefault haiku and subagent slots in Claude Code. Lowest latency.

​
Fire Pass keys

Fire Pass keys (fpk_...) default all slots to glm-latest. The model browser shows Fire Pass-supported models: glm-latest, kimi-fast-latest, and kimi-k2p7-code-fast. Fire Pass keys work with direct Fireworks routing only. Use an Azure API key with --provider azure.
Codex does not support Fire Pass keys yet. Use a standard Fireworks API key (fw_...) with Codex.

​
Migration from earlier syntax

FireConnect uses harness-first commands. If you have older docs or scripts from pre-0.5.0 releases, update them:
BeforeAfter
fireconnect onfireconnect claude on
fireconnect offfireconnect claude off
fireconnect statusfireconnect claude status
fireconnect listfireconnect claude status
fireconnect set --main <id>fireconnect claude on --main <id>
fireconnect resetfireconnect claude model reset
fireconnect on --harness opencodefireconnect opencode on
fireconnect model listfireconnect <harness> model list

​
Source

FireConnect is open source: github.com/fw-ai/fireconnect