Consult every oracle from one temple.
Alexa gathers the leading AI models behind a single wallet and a single API. Prices are posted before you ask, budgets are yours to set, and every balance settles on Robinhood Chain.
Chain 4663 Sign in with any EVM wallet. No seed phrase, ever.
390 models from 52 providers · one API
- GPT81
- Qwen42
- Gemini39
- Claude24
- Mistral21
- GLM20
- DeepSeek18
- Llama14
- Kimi10
- MiniMax9
- Grok9
- Hunyuan7
- Nemotron6
- Aion6
- Seed6
- Command5
- MiMo5
- Nova5
- Sonar5
- Mercury3
- Step3
- Ling3
- Sakana3
- Hermes3
- Solar2
- Reka2
Choose the mind
the question deserves.
A short task does not need a frontier model, and a hard one should not get a cheap guess. Compare every model by tier, context and price per million tokens, then use it in the Sanctum or through the API.
Claude Opus 5.5
anthropic/claude-opus-5.5Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...
- Input / 1M
- $4.20
- Output / 1M
- $21.00
- Context
- 1M
- Max output
- 128K
Most used on Alexa
Ranked by requests billed through Alexa since launch.
Fresh from the workshops
A marketplace of models,
kept in order.
Developers, teams and autonomous agents meet in the same place: one balance, one set of keys, and limits that hold no matter which model or tool sends the request.
Inference
- Leading providers behind one endpoint
- Streaming, tools and structured output
- Switch models per request
Treasury
- Prepaid balance in USD
- Funded with ETH on Robinhood Chain
- Every request itemised
Limits
- Daily spending cap per wallet
- Enforced across all keys
- Agents cannot overspend
Integration
- Drop-in for OpenAI SDKs
- Works with coding agents and IDEs
- Revocable keys, up to ten per wallet
Change the base URL. Keep your code.
Alexa speaks the OpenAI chat format, so existing SDKs, agents and tools work after a one-line change. Create a key in the Sanctum and every request is billed to your wallet.
Read the API guidecurl https://www.alexalabshq.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $ALEXA_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "anthropic/claude-opus-5.5", "messages": [{"role": "user", "content": "Know thyself."}]}'Fees return to
those who hold.
A share of every platform fee funds daily inference credits for ALEXA holders. The token is not a ticket to enter; it is a stake in the temple’s upkeep, paid back in the thing the temple makes.
Protocol fees
Each request carries a 5% fee on top of the provider's reference price.
ALEXA holders
Wallets above the holding threshold on Robinhood Chain qualify automatically.
Daily credits
Credits refill every UTC day and are spent before your own balance, on any model.
- Latest block
- #103,522,452
- Block time
- 0.10 s Average of last 1,000 blocks
- Gas price
- 0.020 gwei
- ETH / USD
- $2,267.35 Coinbase spot
- Requests served
- 8
- Tokens processed
- 45K
- Paid for inference
- $0.10
- Wallets signed in
- 3
- Deposits credited
- 1 0.002 ETH in total
- Treasury balance
- 0.0084 ETH $19.02 · 0x8445…2b90
Asked at
the agora.
How Alexa bills, what it stores and what the token does.
What is Alexa?
Alexa is an inference gateway. It puts 390 models from 52 providers (OpenAI, Qwen, Google, Anthropic, Mistral, Z.ai and more) behind one wallet and one API, with prices shown up front and balances settled on Robinhood Chain.
How do I pay for inference?
Sign in with an EVM wallet, then deposit ETH on Robinhood Chain from the Treasury page. Once the transfer has a few confirmations it is converted to a USD balance at the spot price, and each request is deducted at the model's listed rate.
How is a request priced?
Every model shows its input and output price per million tokens. That is the provider's reference price plus a 5% Alexa fee. You are charged for the tokens a request actually uses; there is no subscription.
Can I cap what I spend?
Yes. Set a daily limit in the Treasury. Once the day’s spend reaches it, the Sanctum and every API key stop accepting requests until 00:00 UTC, or until you raise the limit.
Who sees my conversations?
Conversations in the Sanctum are kept in your browser, not on Alexa’s servers. Prompts are forwarded to the model provider that runs the model you chose, and that provider’s data policy applies to them.
What do ALEXA holders receive?
The ALEXA token has not launched yet. Once it is live on Robinhood Chain, holders above a set threshold will receive daily inference credits funded from platform fees. The threshold and credit amount will be published here at launch.
Do I need ALEXA to use Alexa?
No. Anyone with a wallet and a deposit can use every model. Holding the token only adds the daily credit.
Γνῶθι σεαυτόν.
Then ask the oracle.
Every model you need, one balance to fund them, and a limit you set yourself.