@dsh-external/dsh-vision
Vision for text-only DeepSeek: a view_image tool that answers questions about images via any OpenAI-compatible VLM endpoint (defaults to the free Zhipu glm-4.6v-flash; DashScope, Ollama and others supported)
AI 분석
核心用途是让纯文本模型具备图像理解能力。适合需要让模型分析图片、进行 OCR 的用户,默认支持智谱 glm-4.6v-flash 免费端点,也支持 DashScope 和本地 Ollama。
설치
검증된 bundle이 없거나 호환성 검사에 실패했습니다. 먼저 저장소 설명을 읽어 주세요. 전체 README 읽기 ↗
Configuration(view_image)
| Option | Default | Notes |
|---|---|---|
apiKey | '' | 依次回退 VISION_API_KEY → ZHIPUAI_API_KEY → DASHSCOPE_API_KEY(~/.dsh/.env 或环境变量)。本地端点留空 |
baseURL | https://open.bigmodel.cn/api/paas/v4 | 任意 OpenAI 兼容基址;Ollama:http://localhost:11434/v1 |
model | glm-4.6v-flash | 本机实测配置:DashScope qwen3-vl-flash(快、免费档、OCR 好) |
fallbackModels / maxTokens / timeoutMs / maxImageBytes | 免费链 / 2048 / 60s / 10MB | 见 ~/.dsh/cordis.patch.yml 示例 |