Subscription · live

OpenCode Go
Low-cost lane for open coding models

Works in OpenCode or any agent via OpenCode Go provider. One subscriber per workspace. Includes DeepSeek V4 Flash/Pro, Kimi K3, GLM-5.2, Grok 4.5, GPT-5.6 Luna, and more.

Updated 2026-08-30

#$5 → $10#$12 / 5h#Flash 2× promo#Go ≠ Zen

Highlights

$10

After month 1

$5 first month. Cancel anytime. Top-up / Zen balance optional.

$12

5-hour dollar cap

Also $30/week, $60/month. Request counts depend on model price.

63,300*

Flash / 5h on homepage

2× promo. Docs baseline 31,650. Always footnote.

Go≠Zen

Different product

Zen is the console/login. Go is one paid lane.

What Go is

This is the $10 “Claude Max but open models” search. Be precise about dollar caps vs huge Flash request estimates.

Why it is trending

K3/Grok 4.5 are expensive inside Go (~110–120/5h). Flash is the volume monster. Don’t let the homepage’s 63k steal the FAQ.

Timeline

docs/go

Authoritative limits table.

promo

Flash 2× usage — time-limited on the marketing page.

workspace

Only one member can subscribe Go.

Confirmed vs caution

Confirmed

Price, dollar windows, model list, endpoint, privacy table. DeepSeek ZDR: as of 2026-08-30 fetch, official text still said valid through August 31, 2026, renewed monthly.

Caution

Homepage request counts are estimates. Model list can change. ZDR date will expire.

How to choose

Want cheap open models

Go, and prefer Flash/MiniMax-class ids inside the cap.

Want one frontier model only

Buying official APIs à la carte may be simpler than Go’s mixed menu.

How to use it

/connect → OpenCode Go. Config opencode-go/<id>. Endpoints under opencode.ai/zen/go/v1/…

On QCode

OpenCode Go is a third-party subscription from Anomaly and QCode does not resell it. QCode takes a different route: a pay-as-you-go multi-model API key, with the catalogue on /models currently spanning Claude, GPT, Gemini, DeepSeek, Kimi, GLM and Qwen.

FAQ

Go vs Zen?

Zen = account/console. Go = $10 lane.

Really 63k Flash requests?

Homepage 2× promo estimate. Docs 31,650 / 5h baseline. Dollars are the real cap.

Team seats?

Docs: one Go subscriber per workspace.

Includes 0731 Flash?

Official X said new V4 Flash is in Go and Free. Confirm cutoff in /models.

After limit?

Optional Zen balance, else free models.

Any agent?

They say so via OpenAI-compatible endpoints. Not a promise Claude Code is one-click.

Sources

opencode.ai/go ; opencode.ai/docs/go/ — docs win on numbers.

Sell the dollar cap, not the 63k banner

Engineers should paste docs numbers and footnote homepage 2×.