gaents Dashboard

From zero to a paid answer

Sign up, get a key, and call a model. It's one copy-paste, and every call is a signed row in the ledger.

Prefer clicking? The dashboard signs you up in one click, runs a model for you, and shows your balance, calls and receipts.

Open the dashboard

Statements about the API on this page come from probes run when the page was built (2026-11-19T06:07:31Z) or from the dated check below.

Starter credit: $0.10 per new account, as the API's deploy config sets it (accounts.free_credit_usd, read 2026-11-19T04:39:16Z).

Last end-to-end check (2026-11-19T03:51:19Z): sign-up answered 201 with a starter credit of $0.10 (the API said: “$0.10 starter credit per new account: enough to try hundreds of cheap model calls or 1 minute of premium transcription, so a first call costs a visitor nothing; total exposure capped by accounts.free_credit_budget_usd and the 3/10 s sign-up limit. Adam 2026-11-18 ('let's just get on with it'; earlier 'make your best judgement ... save as admin-user (my) default settings'). Hustle's judgement, unruled; Adam may override.”). The paid call to gaents/glm-flash answered 200 in 0.99 s, and the balance went from $0.100000 to $0.099998. This is a result as of 2026-11-19T03:51:19Z, not a live reading.

1. Get a key

curl -s -X POST https://api.gaents.com/v1/accounts -H 'Content-Type: application/json' -d '{"label":"me"}'
# {"account": "<hex>", "api_key": "...", "free_credit_usd": ...}  (the key is shown once)

2. Ask a model (curl, or the OpenAI SDK)

curl -s https://api.gaents.com/v1/chat/completions -H "Authorization: Bearer $GAENTS_KEY" -H 'Content-Type: application/json' \
  -d '{"model": "gaents/glm-flash", "messages": [{"role": "user", "content": "Hello"}], "max_tokens": 200}'
from openai import OpenAI
client = OpenAI(base_url="https://api.gaents.com/v1", api_key=GAENTS_KEY)
print(client.chat.completions.create(model="gaents/glm-flash", messages=[{"role": "user", "content": "Hello"}]).choices[0].message.content)

gaents/glm-flash: $0.1741 per 1M prompt tokens, $0.5802 per 1M completion tokens (live from /v1/models, 2026-11-19T06:07:16Z). You pay the provider's reported cost plus the margin; the unused part of each quote is refunded in the same request.

Models served now: gaents/claude-haiku · gaents/claude-opus · gaents/claude-sonnet · gaents/deepseek · gaents/deepseek-flash · gaents/fable · gaents/glm-flash · gaents/gpt-oss · gaents/grok · gaents/qwen-27b · gaents/qwen-flash. Which one for your job: routes compared per task, by cost per correct output.

3. Pay without an account (x402)

Leave the key off and the same call answers 402 with a signed quote. Pay it and retry with the proof in X-PAYMENT. Rails offered now: internal_balance, bsv_tx. Discovery is at https://api.gaents.com/.well-known/x402 (answered 200 when this page was built).

curl -i https://api.gaents.com/v1/chat/completions -H 'Content-Type: application/json' -d '{"model": "gaents/glm-flash", "messages": [{"role":"user","content":"Hello"}]}'
# HTTP/1.1 402  ->  accepts[]: price, pay-to, quote id, signed quote
curl https://api.gaents.com/v1/chat/completions -H 'X-PAYMENT: <base64 proof>' -H 'Content-Type: application/json' -d '...same body...'

4. From an agent (MCP)

Endpoint: https://api.gaents.com/mcp (JSON-RPC: initialize, tools/list, tools/call). It answered 200 when this page was built.

curl -s -X POST https://api.gaents.com/mcp -H 'Content-Type: application/json' -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
curl -s -X POST https://api.gaents.com/mcp -H "Authorization: Bearer $GAENTS_KEY" -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"chat","arguments":{"model":"gaents/glm-flash","messages":[{"role":"user","content":"Hello"}]}}}'

5. Check what you paid for

Each call leaves a signed proof: input and output hashes, the model the provider reported, its cost, and your price. See Prove it.