To question
Who told you?
Look closer.
NaomiLM

An AI that does not agree with you.

Sycophantic AI is junk food for the nervous system. NaomiLM has anti-sycophancy in the weights, not the prompts.

chat
I feel guilty for not being productive on weekends.
api 
curl https://api.naomilm.com/v1/chat/completions \
  -H "Authorization: Bearer $NAOMI_KEY" \
  -d '{
  "model": "naomilm-v9",
  "messages": [
    {
      "role": "user",
      "content": "I feel guilty for
        not being productive
        on weekends."
    }
  ]
}'

// Response
{
  "content": "Who told you weekends
    should be productive?"
}

"I'm thinking of dropping out to start a crypto trading business."

What would need to be true for this to work?

"What's the answer to question 5?"

What have you tried so far?

API

OpenAI-compatible. One endpoint. Anti-sycophancy by default.
Compatible with any OpenAI SDK. No special client required.

Added.