opencode-go OpenAI-compatible

Flat-rate subscription gateway from the opencode team — OpenAI-compatible.

Sign up via referral link ↗

opencode Zen Go is a flat-rate subscription endpoint operated by the opencode team — $5 the first month, then $10/month. It exposes a curated open-source catalogue (DeepSeek V4, GLM 5.x, Kimi K2.5/K2.6, MiMo V2.5, MiniMax M2.5/M2.7, Qwen 3.5/3.6) behind a single OpenAI-compatible key issued from opencode.ai. There is no per-token bill: usage draws against dollar-denominated caps — $12 per 5-hour window, $30/week, $60/month — and how many requests you get depends on the model (cheaper ones like MiniMax M2.5 stretch further; GLM-5.1 burns the allowance faster). With Zen credits, enabling “Use balance” lets you continue past the caps.

Strengths

  • Single key for the same catalogue opencode itself uses
  • Cheap flat subscription — $5 first month, then $10/mo, no per-token bill
  • Curated open-source line-up (DeepSeek / GLM / Kimi / MiMo / MiniMax / Qwen)

When to use it

  • Already using opencode and want one bill instead of N provider keys
  • Steady daily use where a $10/mo flat rate beats metered per-token billing
  • Cross-route drift checks vs the same model on direct providers

Subscription plans

PlanPriceQuotaAvailable
Go$10/mo$5 first month · caps: $12/5h · $30/wk · $60/moyes
Notes: opencode Zen Go bills as a flat subscription — $5 the first month, $10/month after — not per token. Usage is capped in dollars ($12 per 5h, $30/week, $60/month); the catalogue's per-model rates (visible via `opencode models --verbose`) only set how fast you draw down those caps. Only one workspace member can hold the Go subscription.
Referral: opencode runs a referral program: sign up via the link above and both you and the referrer earn bonus credits.

Models tested on opencode-go

Speed numbers below are specific to opencode-go's routing and hardware. The same model may appear on other providers' pages with different throughput.

No successful runs recorded for this provider yet.