Which API Provider Should You Use With OpenCode?

Looking for OpenCode plans or a subscription? OpenCode pricing: Go $10, Go Plus $40, Zen and API costs →

Which API provider to connect to OpenCode depends on how you use it: OpenCode Go for the native path and mostly open models, DeepSeek's direct API for light, bursty use, and a cost-optimized routed provider like Standard Compute, which sends each turn to the cheapest model that can do it on a monthly plan from $19, once OpenCode is your daily driver. OpenCode lists Standard Compute natively through models.dev, with the base URL https://api.stdcmpt.com/v1, so switching providers is a config selection, not a setup project. The catch nobody puts in the pricing table: single-vendor plans get reshaped mid-flight (price hikes, new weekly caps), and per-token billing turns one runaway agent loop into a surprise invoice. Pick by which failure mode you can live with.

Light or occasional use

For occasional use, DeepSeek pay-as-you-go is genuinely hard to beat: light OpenCode work lands in single-digit dollars with zero lock-in. OpenCode Go at $10/month is the cheapest subscription, with the usual single-vendor caveats: model-specific allowances, windows, and plan terms that have changed before. Run one representative task on each before committing.

Daily use, cost-optimized

Standard Compute is the cost-optimized, fixed-bill option: one OpenAI-compatible endpoint, smart routing sending routine turns to efficient models (GLM 5.3-flash-class) and hard ones to stronger models, and a plan from $19/month whose price is a hard monthly compute budget. For every model we vet providers for price, uptime and speed, with automatic fallback. Opt-in Europe or US Compliance rules apply per key, team or company, and we don't train on your prompts or store them. No overage charges: requests stop when it is spent. Honest sizing: the $19 tier is for real-but-moderate use; all-day agentic grinding wants a mid tier.

API and subscription options for OpenCode

Prices and allowances can change. OpenCode Go prices were checked against its linked public pages on October 13, 2026, DevPass prices on September 5, 2026; other entries link to current plans without quoting an unverified price. Spot an error? Tell us.

Setup for one model in OpenCode: DeepSeek · Kimi K3 · GLM 5.3 · Qwen 3.8 · GPT · Gemini

OpenCode questions

Is $100 to $150 a seat per month realistic for heavy daily OpenCode use on our own flat inference?

Yes, if routine turns run on efficient models; no, if every turn is pinned to Opus. On Standard Compute the Standard plan is $89 a month (USD) with $95 of compute and Pro is $249 with $269, and the budget is drawn at each model's per-token rate: Claude Opus 5.5 lists at $4 / $20 per million tokens, GLM 5.3 at $1.40 / $4.40 and GLM 5.3 Flash at $0.15 / $0.50. With the model standardcompute, routing sends hard steps to the frontier model and routine ones to the efficient ones, which is what lets one $89 seat carry a full working day; a developer who keeps Opus pinned all day belongs on Pro. A subcontractor can run its own company workspace with one key per developer, separate from the client's Copilot seats, and each key stops at its budget with HTTP 402 instead of billing more.