All coding subscriptions
OpenCode Go
OpenCodeentry$10
Last checked: 2026-07-29
- First limit you hit
- 60 USD of model usage per month
- Equivalent API value
- $60Priced on GLM-5.2
- Value multiple
- 6.00×
- When it runs out
- falls back to your Zen balance after you've reached your usage limits
- Scenario
- API use authorized
Allowances
12USD of model usage per 5 hours
›The vendor's own wording
"OpenCode Go includes the following limits: 5 hour limit — $12 of usage; Weekly limit — $30 of usage; Monthly limit — $60 of usage. Limits are defined in dollar value. This means your actual request count depends on the model you use."Open the source page
30USD of model usage per week
›The vendor's own wording
"Weekly limit — $30 of usage", in the same list, under the caveat the page attaches to all three: "Usage limits may change as we learn from early usage and feedback."Open the source page
60USD of model usage per month
›The vendor's own wording
"Monthly limit — $60 of usage." The page states the intent behind the figure as well: "With Go, you pay $10/month and we aim to give you 6x that in usage."Open the source page
Narrowed for specific models
These are not extra limits. Each one lowers the plan-wide allowance above for a single model, so the plan is worth what its best-covered model can spend, not this figure.
Grok 4.515USD of model usage per month (this model only)
›The vendor's own wording
The price table gives Grok 4.5 a "Usage" of "$15", against "$60" for most rows, under the heading "The estimates are also based on the following prices per 1M tokens and the monthly usage included with each model".Open the source page
Kimi K315USD of model usage per month (this model only)
›The vendor's own wording
The price table gives Kimi K3 a "Usage" of "$15". The page explains the class: "For some models, we haven't had the opportunity to negotiate a discount or host them at a lower cost, either because the model is new or because their public pricing is already discounted."Open the source page
DeepSeek V4 Pro15USD of model usage per month (this model only)
›The vendor's own wording
The price table gives DeepSeek V4 Pro a "Usage" of "$15", priced at "$0.435" input and "$0.87" output per 1M tokens inside Go.Open the source page
mimo-v2.5-pro15USD of model usage per month (this model only)
›The vendor's own wording
The price table gives MiMo V2.5 Pro a "Usage" of "$15", alongside "$60" for the non-Pro MiMo V2.5 in the row above it.Open the source page
Models included
| Model | Capability | Burn rate | |
|---|---|---|---|
| Grok 4.5own cap 15 | frontier | — | |
| GLM-5.2 | untiered | — | |
| GLM-5.1 | untiered | — | |
| Kimi K3own cap 15 | untiered | — | |
| Kimi K2.7 Code | untiered | — | |
| Kimi K2.6 | untiered | — | |
| MiniMax M3 | untiered | — | |
| MiniMax M2.7 | untiered | — | |
| Qwen3.7 Max | untiered | — | |
| Qwen3.7 Plus | untiered | — | |
| Qwen3.6 Plus | untiered | — | |
| DeepSeek V4 Proown cap 15 | untiered | — | |
| DeepSeek V4 Flash | untiered | — | |
| MiMo-V2.5 | untiered | — | Included in Go at "$0.14" input / "$0.28" output per 1M tokens with "$60" of monthly usage. Zen lists only a free promotional variant, so there is no paid per-token row to value this against. |
| MiMo-V2.5-Pro | untiered | — | Included in Go at "$0.435" input / "$0.87" output per 1M tokens, and capped at "$15" of monthly usage rather than $60. Not listed on Zen. |
| Hy3 | untiered | — | Included in Go at "$0.14" input / "$0.58" output per 1M tokens with "$60" of monthly usage. Tencent's Hy3 is not listed on Zen. |
Subscription versus your own API key
This compares two ways of working the same way — in an IDE or CLI, where BYOK tools make a direct API key a real option. It is not a claim that subscriptions are cheaper than the API in general.
| Model | Effective price / 1M tokens | vs API rate | Breakeven |
|---|---|---|---|
| Grok 4.5frontier | $0.92 | 0.67 | 7.2M |
| GLM-5.2untiered | $0.17 | 0.17 | 9.8M |
| GLM-5.1untiered | $0.17 | 0.17 | 9.8M |
| Kimi K3untiered | $1.77 | 0.67 | 3.8M |
| Kimi K2.7 Codeuntiered | $0.14 | 0.17 | 12.3M |
| Kimi K2.6untiered | $0.13 | 0.17 | 12.5M |
| MiniMax M3untiered | $0.04 | 0.17 | 39.9M |
| MiniMax M2.7untiered | $0.04 | 0.17 | 39.9M |
| Qwen3.7 Maxuntiered | $0.3 | 0.17 | 5.5M |
| Qwen3.7 Plusuntiered | $0.05 | 0.17 | 32.3M |
| Qwen3.6 Plusuntiered | $0.08 | 0.17 | 20.1M |
| DeepSeek V4 Prountiered | $0.63 | 0.67 | 10.6M |
| DeepSeek V4 Flashuntiered | $0.01 | 0.17 | 116.5M |
Usage terms
Quoted from the vendor's own terms. We do not give legal advice.
The docs publish the programmatic path rather than restricting it: "You can also access Go models through the following API endpoints." Nothing contradicts it, because there is nothing to contradict it with — opencode.ai/terms, /legal and /privacy all 404 as of 2026-07-29. The authorization rests on the documented endpoints and on nothing else.
Change history
- 2026-07-29Recorded from the seed file
- 2026-07-29Recorded from the seed file