One OpenAI-compatible API
Point an existing OpenAI client at the SynapseX base URL and keep your code. Chat completions, streaming over SSE and a /v1/models catalog behave the way your stack already expects.
SynapseX API Platform
SynapseX: LLM API provider with one API, one wallet, every model
One OpenAI-compatible API gives your application access to SynapseX proprietary models and frontier models alike, billed per token against a single credit balance from $0.10 per million input tokens.
Call the OpenAI-compatible API in one request
Create an API key in the console, export it, and send a standard chat completion. The response shape, streaming semantics and error envelope match what OpenAI-compatible clients already parse.
curl https://platform.synapsex.ai/api/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $SYNAPSEX_API_KEY" \
-d '{
"model": "synapsex-atlas-7b-v1",
"messages": [{"role": "user", "content": "Hello Atlas!"}]
}'What the LLM provider gives your team
Per-token pricing from one wallet
Pay-as-you-go AI API billing: input and output tokens are metered separately and deducted from a single prepaid credit balance. No subscription, no monthly minimum, no per-model contract.
AI API keys with scopes
Issue scoped keys per workspace, revoke them independently, and keep request and token rate limits that scale automatically with the credit balance you hold.
Usage and cost analytics
Every request is metered. Break spend down by model, key and workspace so cost attribution never depends on a spreadsheet.
AI agent runtime API
Deploy an agent once and invoke it over HTTP the same way you call a model, with the same keys, the same wallet and the same usage records.
Python, Node and Go AI SDKs
A first-party Python SDK on PyPI, plus copy-paste Node and Go clients that reuse the OpenAI-compatible wire format against the same base URL.
View the client SDKsPay-as-you-go AI API pricing
11 models are live today, each with a published per-token rate. Credits are consumed proportionally to usage, unused credits roll over, and rate-limit tiers rise with the balance you hold rather than with a plan you have to negotiate.
Compare per-token pricing for every model