Alibaba Coding Plan
Qwen and partner models on an Alibaba Cloud Coding Plan. Bring your own key; nothing is billed by us.
Esta página está escrita en inglés. Los términos que explica son los que se buscan en inglés, y un término técnico traducido es un término distinto.
What the harness already knows
| Provider id | qwen-coding-plan |
| Base URL | https://coding-intl.dashscope.aliyuncs.com/v1 |
| Protocol | OpenAI-compatible chat completions |
| API key | ALIBABA_CODING_PLAN_API_KEY |
| Model discovery | No — models come from the catalogue and from what you configure. |
Connecting it
The endpoint above is built in, so there is nothing to configure. Put the key in the environment and the provider appears in the model picker:
export ALIBABA_CODING_PLAN_API_KEY="..."Keys are issued by Alibaba Coding Plan — where to get one.
Models on Alibaba Coding Plan
The newest models Alibaba Coding Plan serves that can call tools, from the models.dev snapshot OHarness ships.
| Model id | Context / output | $/M in / cached / out | Reasoning |
|---|---|---|---|
qwen3.7-plus | 1M / 64K | 0 / 0 / 0 | Yes |
qwen3.7-max | 1M / 66K | 2.5 / 0.5 / 7.5 | Yes |
qwen3.6-flash | 1M / 66K | 0.1875 / — / 1.125 | Yes |
qwen3.6-plus | 1M / 66K | 0 / 0 / 0 | Yes |
qwen3.5-plus | 1M / 66K | 0 / 0 / 0 | Yes |
MiniMax-M2.5 | 197K / 25K | 0 / 0 / 0 | Yes |
glm-5 | 203K / 16K | 0 / 0 / 0 | Yes |
qwen3-coder-next | 262K / 66K | 0 / 0 / 0 | No |
kimi-k2.5 | 262K / 33K | 0 / 0 / 0 | Yes |
qwen3-max-2026-01-23 | 262K / 33K | 0 / 0 / 0 | No |
List prices in USD per million tokens, read from models.dev on 2026-09-19. Alibaba Coding Plan changes its prices and models; its own pricing page is the authority. Start a session on the newest:
oharness -m qwen-coding-plan/qwen3.7-plusWorth knowing
A plan endpoint: its own env var, because a plan key sent to the pay-as-you-go host (or the reverse) is refused or billed per token.
Pointing it somewhere else
A regional host, a gateway, or a server of your own. Overrides in ~/.oharness/config.json are field level, so changing the endpoint does not mean restating the rest of the provider — this is the whole block, shown holding the default so the shape is clear:
{
"providers": {
"qwen-coding-plan": {
"baseUrl": "https://coding-intl.dashscope.aliyuncs.com/v1"
}
}
}What you get either way
The same agent loop, the same tools, the same MCP servers and the same permission rules, whichever provider is underneath — and switching between them mid-session converts nothing, because history is never stored in a vendor’s shape. What an agent harness is covers why that is the hard part, and the providers reference covers the catalogue itself.