Skip to content
@Floe-Labs

Floe Labs

Openrouter for voice AI. Finance function for Voice AI. Unified billing ledger. One key. Every vendor. Every model.

Floe Labs

Floe Labs

One key for your voice agent's entire bill — LLM, voice, telephony, data. One endpoint to pay across any LLM or voice vendor API services, with programmable, context-aware budgets. Walletless.

Website · Docs · Dashboard · 𝕏 @FloeLabs

npm pypi license


Start free. $3 Welcome Credit — 300 API credits on signup, no card, no wallet. Your agent makes its first paid API call in minutes. Get started →

Start building — pick your path

Let your coding agent set Floe up, or wire it yourself:

Path One line
Agent — Claude Code / Cursor does the setup paste: Read https://dev-dashboard.floelabs.xyz/agents.md and set up Floe for this project.
Skill — install the Floe agent skill npx skills add floe-labs/agent-skills
MCP — hosted MCP server (65 tools) claude mcp add --transport http floe https://mcp.floelabs.xyz/mcp
NPM — the CLI + SDK npm i -g floe-agent

Set up with your AI tools → · Agent skill →

Why Floe

A token router meters your LLM spend — ~40% of a voice call. Floe meters 100%: transcription, speech, telephony, search, and reasoning, on one key.

Your agent calls a dozen paid APIs — LLMs, voice, search, data. That's a dozen accounts, a dozen prepaid balances, a dozen keys, and no unified way to see or govern what your agent spends. Agents overspend, loop, and stall.

Floe gives your agent a budget, not a balance:

  • One endpoint, 2,000+ LLM inference + vendor API services. Pay any vendor through Floe. No per-vendor accounts or keys.
  • Programmable spend controls. Per-call caps, daily limits, allowed destinations — set at the vendor, agent, or team level, time-bound. Enforced before money moves.
  • Budget-aware routing. Every response carries result.budgetAdvisory — how close the agent is to its tightest cap. Your agent reads it and downgrades, finishes the task, or hard-stops before it overspends, instead of failing at the cap.
  • Walletless. Email + a funding source. We provision wallets in the background — no MetaMask, no seed phrase, no gas. The stablecoin rails are invisible.
  • Real-time visibility. Every call is a typed receipt: target, amount, status, time. Reconcile, alert, or revoke from the dashboard.

Quickstart — one voice turn, one key

A single spoken turn spends across speech-to-text, an LLM, and text-to-speech. Same key, same task budget for all three — a token router only ever sees the middle step.

# 1 · Transcribe (Deepgram) — through the Floe proxy
curl -X POST https://credit-api.floelabs.xyz/v1/proxy/fetch \
  -H "Authorization: Bearer $FLOE_API_KEY" \
  -H "Content-Type: application/json" \
  -H "X-Floe-Task-Id: call-8842" \
  -d '{"url":"<DEEPGRAM_STT_ENDPOINT>","method":"POST","body":"{...audio...}"}'

# 2 · Reason (any LLM) — keyless: only your Floe key, same ledger
curl -X POST https://credit-api.floelabs.xyz/v1/chat/completions \
  -H "Authorization: Bearer $FLOE_API_KEY" \
  -H "Content-Type: application/json" \
  -H "X-Floe-Task-Id: call-8842" \
  -d '{"model":"openai/gpt-4o","messages":[{"role":"user","content":"Book Friday at 2pm."}]}'

# 3 · Speak (ElevenLabs) — through the Floe proxy
curl -X POST https://credit-api.floelabs.xyz/v1/proxy/fetch \
  -H "Authorization: Bearer $FLOE_API_KEY" \
  -H "Content-Type: application/json" \
  -H "X-Floe-Task-Id: call-8842" \
  -d '{"url":"<ELEVENLABS_TTS_ENDPOINT>","method":"POST","body":"{...text...}"}'

All three legs share X-Floe-Task-Id, so one task budget caps the whole conversation — STT, LLM, and TTS together, on one ledger. Each response returns its own cost in X-Floe-Payment-Amount (e.g. 0.0125), so you can meter spend per call and per task.

MCP (zero install) — add to Claude Code / Cursor / Claude Desktop:

claude mcp add --transport http floe https://mcp.floelabs.xyz/mcp \
  --header "Authorization: Bearer YOUR_FLOE_KEY"

Prefer an SDK? npm install floe-agent or pip install floe-agentkit-actions — see Repos.

Voice Stack quickstart →

Already have a pipeline?

Don't rebuild anything. Route just the LLM leg — change the base URL and the key:

from openai import OpenAI

client = OpenAI(
    base_url="https://credit-api.floelabs.xyz/v1",  # was https://api.openai.com/v1
    api_key=os.environ["FLOE_API_KEY"],             # your Floe key — not an OpenAI key
)
# every call now bills to your Floe balance, under your spend caps

OpenAI-compatible, so it works from any SDK. Route one leg or your whole voice bill → Add Floe to your existing pipeline.

Vendor Marketplace — 2,000+ vendor API services, one key

Category Services
Compute Venice AI · OpenAI · Anthropic · Google Gemini · z.ai · Kimi
Voice Deepgram · ElevenLabs · Venice AI · Floe Phone via Twilio (live)
Image Venice AI (generation, upscale, edit, background removal)
Web Firecrawl (search + scrape)
Search Exa · Parallel AI · Tavily
Browser Hyperbrowser · Browserbase · Anchor Browser
Agent tools AgentMail · Pinata · PostalForm

Browse the full directory →

Works with your framework

Framework Status How
Coinbase AgentKit GA Native — floeActionProvider
LangChain GA getLangChainTools adapter
Vercel AI SDK GA getVercelAITools adapter
Claude / Cursor GA floe-mcp-server
CrewAI Beta via MCP server
OpenAI Agents SDK Preview MCP fallback; native adapter in progress
Plain HTTP/REST GA anything that speaks HTTP

Repos

Repo What it does Install
agentkit-actions TypeScript SDK — wallet, x402 payments, spend controls, agent awareness npm install floe-agent
agentkit-actions-py Python SDK — full parity pip install floe-agentkit-actions
floe-mcp-server MCP server for Claude, Cursor, any MCP agent Setup
floe-cookbook Runnable end-to-end agents, including voice git clone
eve-floe Reference voice agent built on Floe git clone
floe-guard Local budget guardrail — hard-stops a runaway agent before it overspends pip install floe-guard

Roadmap

Floe is shipping the spend layer first, and building the credit layer on top of it.

  • Working capital / credit linesin development. Borrow against deposits and, later, against your agent's usage and receivables.
  • Portable credit & trust recordin development. Every transaction builds the behavioral data that will let agents be underwritten without re-running diligence.

Want early access to credit features? Talk to us →


Built by operators from Airwallex, Western Union, eBay, Kado, Transak. hello@floelabs.xyz

Pinned Loading

  1. floe-mcp-server floe-mcp-server Public

    TypeScript 4 1

  2. floe-cookbook floe-cookbook Public template

    Agents and demos for the Floe finops for voice builders, including 15 LLM model providers + 2,000+ vendor API services

    TypeScript 3 1

  3. floe-guard floe-guard Public

    Open-source unified billing for Voice AI builders and agents — stop before a runaway loop burns your bill. Local, framework-agnostic, no account required.

    Python 43 13

  4. eve-floe eve-floe Public template

    Spend controls for Vercel Eve agents — hard budget caps + budget-aware tapering via Floe (template + MCP connection).

    TypeScript 3

  5. floe-vercel-ai-starter floe-vercel-ai-starter Public template

    Governed Vercel AI SDK agent starter — hard budget caps out of the box via Floe, the billing layer for voice AI. Clone, deploy, ship.

    TypeScript 2

  6. floe-crewai-starter floe-crewai-starter Public template

    Governed CrewAI agent starter — hard budget caps out of the box via Floe, the billing layer for voice AI. Clone, deploy, ship.

    Python 1

Repositories

Showing 10 of 13 repositories

Top languages

Loading…

Most used topics

Loading…