Launching today
Synap402
OpenAI-compatible API for open models at near-cost prices
1 follower
OpenAI-compatible API for open models at near-cost prices
1 follower
One OpenAI-compatible endpoint for open-weight models (Qwen3 Coder, gpt-oss-20b, Gemma 4, Qwen3-VL) priced near cost: gpt-oss-20b is $0.017 in / $0.085 out per 1M tokens. Free tier: 100K tokens/day. Agents can pay in USDC over x402, no account.
Free Options
Launch Team
Render Sandboxes: a computer for every agentIsolated compute, persistent state and a real filesystem
Promoted
Hi Product Hunt! We built Synap402 because most of what agents and coding tools send is small, repetitive work that does not need a frontier model.
Synap402 is one OpenAI-compatible endpoint (https://synap402.com/v1) for open-weight models. Change base_url and key in any OpenAI SDK, Claude Code, Codex, Cline or Aider.
Live prices per 1M tokens (in/out): gpt-oss-20b $0.017/$0.085, Qwen3 Coder 30B $0.0675/$0.27, Gemma 4 26B $0.04/$0.21. Batch is half.
Two things we care about: privacy (we never train on your prompts, and we don't keep the text of your API prompts or replies), and x402 payments, so an agent with a USDC wallet can top up on its own with no account.
There is a free tier (100K tokens a day on Qwen3 Coder, no card). We would love feedback on which models you want next.