Decentralized inference for
every model.

One API across OpenAI, Anthropic, Gemini, DeepSeek, FLUX and more, served by a global network of GPU nodes and settled per token in USDT, ETH or SOL.

Start buildingSRead the docs

OpenAI-compatible
USDT / ETH / SOL
120+ models

Modelclaude-sonnet-5.5

Nodeus-east-4 · H100

Latency182 ms TTFT

1,284,033,120

tokens served today

Settled

0.0041 USDT

Latency182 ms TTFT

[Trusted by the teams at]

Parsec

Vectorly

Nimbus Labs

Octane

Lattice

Moonbase

Helio

Quorum

Tessera

Driftwood

“We moved 40% of our agent traffic to Topo in a week. Same models, a third of the bill, and we settle in USDT instead of juggling five vendor invoices.”

Maya Chen, Head of AI Infrastructure at Parsec

MC

Parsec

Tokens served, last 30 days: 0.0B
Models on the network: 0+
Median time to first token: 0ms
Settled on-chain to date: $0.0M

[Live trace]

Watch a request settle.

Every call is routed, verified and paid for in one round trip. This is the trace your agent sees.

OpenAI-compatible
Change the base URL, keep your SDK.
Verified outputs
Attestors re-run a sample of every node's work.
On-chain receipts
Every token has a tx hash.

Read the API docs

topo · zsh

live

$topo run --model claude-sonnet-5.5 "Summarize the Q3 board deck and draft three follow-ups"

[The network]

Built for every layer of the inference stack

[Developers]

One API for every model

Swap your base URL and keep the OpenAI SDK. Frontier and open models, one key, one bill.

Read the quickstart

[Agents]

Wallets that pay per token

Give every agent its own spend limit and a USDT balance. No cards, no invoices, no humans in the loop.

Agent wallets

[Node operators]

Earn for every token you serve

Plug in H100s or a single 4090. Stake, serve verified inference, get paid daily on-chain.

Run a node

[The network]

Built for every layer of the inference stack

[Developers]
One API for every model
Swap your base URL and keep the OpenAI SDK. Frontier and open models, one key, one bill.
Read the quickstart
[Agents]
Wallets that pay per token
Give every agent its own spend limit and a USDT balance. No cards, no invoices, no humans in the loop.
Agent wallets
[Node operators]
Earn for every token you serve
Plug in H100s or a single 4090. Stake, serve verified inference, get paid daily on-chain.
Run a node

cURL

curl https://api.topolabs.ai/v1/chat/completions \-H "Authorization: Bearer tp_live_8f3a…" \-H "Content-Type: application/json" \-d '{"model": "claude-sonnet-5.5","messages": [{ "role": "user", "content": "Explain zk proofs in 3 lines" }],"pay_with": "usdt"}'

Modelclaude-sonnet-5.5

Nodeeu-west-2 · B200

[Under the hood]

Topo routes every request to the best node, verifies the output, and settles the bill in seconds.

[Inference receipt]

claude-sonnet-5.5

Anthropic

-0.0041 USDT

Chain: Base
Tx hash: 0x51c4…9ab2
Tokens in / out: 812 / 249
Node: eu-west-2
Settled at: 2026-10-16 09:23:16

[API key]

tp_live_8f3a••••••••••••••••••••c21e

Latency

0

ms

[Models]

Every model. One price list.

Frontier labs and open weights, side by side. Prices per million tokens, settled in USDT.

OpenAI

GPT-6.1 Sol

ChatContext 1M

$2.00 in · $10.00 out

/ 1M tokens

OpenAI

GPT-6 Astra

ReasoningContext 1M

$10.00 in · $50.00 out

/ 1M tokens

Anthropic

Claude Fable 5.1

ReasoningContext 1M

$10.00 in · $50.00 out

/ 1M tokens

Anthropic

Claude Opus 5.5

ReasoningContext 1M

$4.00 in · $20.00 out

/ 1M tokens

Anthropic

Claude Sonnet 5.5

ChatContext 1M

$2.00 in · $10.00 out

/ 1M tokens

Google

Gemini 4 Argon

ReasoningContext 1M

$2.00 in · $10.00 out

/ 1M tokens

DeepSeek

DeepSeek V4 Pro

ReasoningContext 1M

$1.32 in · $3.96 out

/ 1M tokens

Muse Spark 1.3

ChatContext 1M

$1.25 in · $4.25 out

/ 1M tokens

Alibaba

Qwen3.8-Max

ReasoningContext 1M

$2.00 in · $6.00 out

/ 1M tokens

Mistral

Mistral Large 4

ChatContext 256k

$1.36 in · $4.18 out

/ 1M tokens

xAI

Grok 4.7

ChatContext 500k

$2.00 in · $6.00 out

/ 1M tokens

OpenAI

GPT Image 2

Image

$0.053 / image

Black Forest Labs

FLUX.2 [pro]

Image

$0.03 / image

Google

Veo 3.1

Video

$0.40 / second

OpenAI

gpt-4o-transcribe

Audio

$0.006 / minute

OpenAI

text-embedding-3-large

EmbeddingContext 8k

$0.13 in

/ 1M tokens

120+ models · prices update hourlyFull price list

[How it works]

From request to receipt in under a second.

Your appOpenAI SDK · REST

Topo Routerapi.topolabs.ai

Node network40+ regions · staked

SettlementBase · Solana

01
Your app
OpenAI-compatible SDKs, REST, or an agent framework. Change one line: the base URL.
- Chat
- Embeddings
- Images
- Video
02
Topo Router
Authenticates the key or wallet, picks the fastest eligible node, enforces spend limits.
- Auth
- Routing
- Rate limits
- Verification
- Settlement
03
Node network
Staked GPU operators in 40+ regions serve the request and attest to the output.
- H100 · B200 · 4090
- 40+ regions
- Staked
04
Settlement
The bill is split between the node and the network and settled on-chain with a receipt.
- USDT
- ETH
- SOL
- Base · Solana

Works with your stack:

OpenAI SDK·Vercel AI SDK·LangChain·LlamaIndex·CrewAI·Mastra·Python·TypeScript

OpenAI SDKVercel AI SDKLangChainLlamaIndexCrewAIMastraPythonTypeScriptOpenAI SDKVercel AI SDKLangChainLlamaIndexCrewAIMastraPythonTypeScript

[Dashboard]

Built for teams. Shared wallets. Spend limits per key. Usage by model, node and region. Audit log for every token. Founder-led support.

Open the dashboard

Free $5 credit. No card required.

app.topolabs.ai

Parsec · Max

Usage · last 30 days

1 Oct to 30 Oct 2026

Wallet balance

2,391.08USDT

30-day spend

$1,206.14

Requests

1,755,458

Daily spend

USDT / day

1 Oct15 Oct30 Oct

Usage by model

share of tokens

claude-sonnet-5.542%
gpt-6.1-sol23%
deepseek-v4-pro17%
muse-spark-1.311%
flux.2-pro7%

API keys

4 active

Key	Spend / limit	Requests	Status
prod-agents	$612.40 / $1,000	1,204,331	live
research	$318.90 / $500	88,120	live
staging	$142.60 / $250	410,993	live
evals	$132.20 / $200	52,014	live

[Pricing]

Pay per token. Settle in crypto.

No seats, no minimums. Top up with USDT, ETH or SOL, or pay in TOPO and save 20%.

Pay as you go

$0

/mo

Per-token pricing on every model. Start with $5 of free credit.

Start free

Every model on the networkFrontier + open weights
OpenAI-compatible APISDKs for Python & TypeScript
2 API keysSpend limits per key
Pay in USDT, ETH, SOLOn-chain receipts
Community supportDiscord + docs

Builder

$29

/mo

For builders shipping their first agents to production.

Start free trial

Everything in Pay as you go, plus:

$85 of inference creditIncluded every month
10% off every tokenAcross all models, always
10 API keysPer-key budgets and alerts
Priority routingLower queue time at peak
Email support48h response

Max

Enterprise

Custom

Private nodes, custom contracts and invoicing in USDT or fiat.

Contact sales

Everything in Max, plus:

Private node poolsBring your own GPUs
Invoicing in USDT or fiatNet-30, PO-based
SSO + SCIMOkta, Entra, Google
Custom verificationAttestor sets you choose
24/7 supportNamed engineer

Prices in USD, settled in USDT. Inference credit never expires.

Run a node.

Node operators

Earn up to 85% of every token your GPU serves, paid daily in USDT. H100s to a single 4090.

85%revenue share

Become an operator

[FAQ]

Frequently Asked Questions

Everything teams ask before switching their base URL. Can't find yours?

Write to support@topolabs.ai

Yes. Point the SDK at https://api.topolabs.ai/v1 and keep the same request shapes. Chat, embeddings, images, audio and streaming all work unchanged.

Run inference on every model, settle in seconds.

Start with $5 of free credit. No card, no invoice. Just a wallet.

Start buildingSTalk to us

Decentralized inference forevery model.

Watch a request settle.

OpenAI-compatible

Verified outputs

On-chain receipts

Built for every layer of the inference stack

One API for every model

Wallets that pay per token

Earn for every token you serve

Built for every layer of the inference stack

One API for every model

Wallets that pay per token

Earn for every token you serve

Topo routes every request to the best node, verifies the output, and settles the bill in seconds.

Every model. One price list.

GPT-6.1 Sol

GPT-6 Astra

Claude Fable 5.1

Claude Opus 5.5

Claude Sonnet 5.5

Gemini 4 Argon

DeepSeek V4 Pro

Muse Spark 1.3

Qwen3.8-Max

Mistral Large 4

Grok 4.7

GPT Image 2

FLUX.2 [pro]

Veo 3.1

gpt-4o-transcribe

text-embedding-3-large

From request to receipt in under a second.

Your app

Topo Router

Node network

Settlement

Built for teams. Shared wallets. Spend limits per key. Usage by model, node and region. Audit log for every token. Founder-led support.

Usage · last 30 days

Pay per token. Settle in crypto.

Pay as you go

Builder

Max

Enterprise

Run a node.

Frequently Asked Questions

Is Topo compatible with the OpenAI SDK?

Which currencies can I pay with?

How are outputs verified on a decentralized network?

Where does my data go?

What latency should I expect?

How do node operators get paid?

Run inference on every model, settle in seconds.

Decentralized inference for
every model.