Any OpenAI SDK works with the base URL set to the Mothpoint and your key as the API key. Keys start with moth_, and every reply reports its charge in the moth-charged-micros header or the stream's last usage chunk.
What a workload costs
At $0.75 / $3.75 per 1M tokens, input / output
With MothAt listYou save
Per call$0.003375$0.006750$0.003375
Per day$0.3375$0.6750$0.3375
Over 30 days$10.125$20.250$10.125
The Moth price is what your balance is charged, settled to the micro-dollar on each call. Over 30 days that is 50% below list. A token is about four characters of English.
Your use
Last 30 days
Parameters
What POST /v1/chat/completions takes for this model
modelstringRequired
gemini-4-hydrogen
messagesarrayRequired
The conversation, oldest first, in the OpenAI format (or Anthropic's on /v1/messages).
max_tokensinteger
Most tokens to write, within the 1.04M-token context. The hold is set from it.
temperaturenumber
0 to 2.
streamboolean
Server-sent events; the final chunk reports usage and the charge.
toolsarray
Function tools, passed to the model as written.
Similar chat models
19 chat models among 48 from 13 makers, nearest in price firstSee every chat model