[Notice]: v0.2.14-v0.3.1 do not support coding plan yet / 注意：v0.2.14-v0.3.1 暂不支持 coding plan

[qin-ctx]

Notice / 注意事项

OpenViking versions v0.2.14 through v0.3.1 currently do not support coding plan.

OpenViking v0.2.14 到 v0.3.1 版本当前 不支持 coding plan。

The reason is that this scenario depends on the model-side Responses API, and compatibility for that path is not available yet.

原因是该场景依赖模型侧的 Responses API，而当前这一调用路径的兼容性尚未完成。

We plan to add compatibility in a future release.

后续版本会补齐这部分兼容能力。

Affected Versions / 受影响版本

• v0.2.14 to v0.3.1

Impact / 影响

• coding plan is not supported on these versions

• Other scenarios are not covered by this notice

• 这些版本下 coding plan 当前不可用

• 本说明仅针对该场景，不代表其他场景不可用

Workaround / 临时处理方式

If you need coding plan for now, please use 0.2.13 for both the plugin side and the OV service side.

如果当前需要使用 coding plan，请将插件侧和 OV 服务侧都先回退到 0.2.13。

Plugin install / 插件安装：

npm install -g openclaw-openviking-setup-helper@latest && ov-install -y --version 0.2.13

OV service rollback / OV 服务回滚：

pip install openviking==0.2.13 --force-reinstall

Additional Notes / 补充说明

• This is a compatibility notice, not a takedown announcement

• Future versions will add support for coding plan on the model-side Responses API path

• 这是兼容性注意事项，不是下线公告

• 后续版本会支持模型侧 Responses API 路径下的 coding plan

[qin-ctx]

Volcengine / Doubao model-side Responses API test script:

import json
import os

from volcenginesdkarkruntime import Ark

API_KEY = os.environ["ARK_API_KEY"]
BASE_URL = os.getenv("ARK_API_BASE", "https://ark.cn-beijing.volces.com/api/v3")
MODEL = os.getenv("ARK_MODEL", "your-ark-endpoint-or-model")
PROMPT = "请用 Python 写一个二分查找函数，并说明时间复杂度。"

client = Ark(api_key=API_KEY, base_url=BASE_URL)

response = client.responses.create(
    model=MODEL,
    input=[
        {"role": "user", "content": PROMPT},
    ],
    thinking={"type": "disabled"},
)


def extract_reply_text(resp) -> str:
    text_parts = []
    for item in getattr(resp, "output", []) or []:
        if getattr(item, "type", None) != "message":
            continue
        for block in getattr(item, "content", []) or []:
            block_type = getattr(block, "type", None)
            if block_type in {"output_text", "text"}:
                text_parts.append(getattr(block, "text", "") or "")
            elif hasattr(block, "text"):
                text_parts.append(block.text or "")
    return "".join(text_parts).strip()


print("response_id:", getattr(response, "id", ""))
print("assistant_reply:\n")
print(extract_reply_text(response) or "(no assistant text found)")

usage = getattr(response, "usage", None)
if usage is not None:
    print("\nusage:")
    print(
        json.dumps(
            {
                "input_tokens": getattr(usage, "input_tokens", 0),
                "output_tokens": getattr(usage, "output_tokens", 0),
                "total_tokens": getattr(usage, "total_tokens", 0),
            },
            ensure_ascii=False,
            indent=2,
        )
    )

Notes:

• install dependency: pip install volcenginesdkarkruntime
• set ARK_API_KEY to your Volcengine ARK API key
• set ARK_MODEL to your endpoint ID or model name
• coding plan is not supported in v0.2.14 to v0.3.1 yet
• temporary workaround:

Plugin install / 插件安装：

npm install -g openclaw-openviking-setup-helper@latest && ov-install -y --version 0.2.13

OV service rollback / OV 服务回滚：

pip install openviking==0.2.13 --force-reinstall

[qin-ctx]

fixed. #1137

[pawellii]

@qin-ctx 这个coding plan指的是火山引擎的吧 模型提供商改为openai貌似就可以了 大佬 可以这么用吗？

"provider": "openai",
"api_base": "https://ark.cn-beijing.volces.com/api/coding/v3",

[qin-ctx]

@qin-ctx 这个coding plan指的是火山引擎的吧 模型提供商改为openai貌似就可以了 大佬 可以这么用吗？

"provider": "openai",
"api_base": "https://ark.cn-beijing.volces.com/api/coding/v3",

能跑过上面的用例吗

[qin-ctx]

更新到 0.3.2 已恢复
npm install -g openclaw-openviking-setup-helper@latest && ov-install -y --version 0.3.2