qwen-omni-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| QWEN_MODEL | No | Model id (multimodal) | qwen3.7-plus |
| DASHSCOPE_API_KEY | Yes | Bailian API key | |
| DASHSCOPE_BASE_URL | No | OpenAI-compatible endpoint | https://dashscope.aliyuncs.com/compatible-mode/v1 |
| QWEN_REQUEST_TIMEOUT | No | Per-request timeout in seconds | 300 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| analyze_videoA | Watch and analyze a video using Qwen3.8-Max (native multimodal). Use this whenever you need to see a video you cannot view natively. The model reads the video natively — no client-side frame extraction. Pass a public URL (http/https) or a local file path; local files are sent inline as a base64 data URL (25MB guardrail). |
| analyze_imageA | View, read, or analyze an image using Qwen3.8-Max (native multimodal). Use this whenever you need to see an image you cannot view natively (e.g., your file reader returns '[Unsupported Image]'). Pass a public URL (http/https) or a local file path; local files are sent inline as a base64 data URL (25MB guardrail). |
| analyze_audioA | Listen to and analyze an audio file using Qwen3.5-Omni (qwen3.5-omni-plus, native audio understanding). Use this whenever you need to hear audio you cannot process natively. Pass a public URL (http/https) or a local file path; local files are sent inline as base64 (25MB guardrail, mp3/wav/flac/ogg/m4a/aac). |
| analyze_audio_videoA | Watch and listen to a video (visuals AND its audio track) using Qwen3.5-Omni (qwen3.5-omni-plus, native audio+video understanding). Use this when what is said or heard in the video matters. Pass a public URL (http/https) or a local file path; local files are sent inline as a base64 data URL (25MB guardrail). |
| check_endpoint_statusA | Check the configured Bailian endpoint, model, and timeout. The API key is redacted in the output. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Most tools are clearly distinct: image, audio, and video analysis have separate tools. The only potential confusion is between analyze_video and analyze_audio_video, but the descriptions clarify that analyze_audio_video explicitly handles audio tracks as well. check_endpoint_status is unrelated and unambiguous.
The primary tools follow a consistent analyze_<media_type> pattern (analyze_video, analyze_image, analyze_audio, analyze_audio_video). The utility tool check_endpoint_status deviates from this pattern but still follows the verb_noun structure. Overall, the naming is predictable and easy to navigate.
With 5 tools, the server is well-scoped for its purpose of multimodal analysis. It covers all major media types (image, audio, video, audio+video) and includes a utility check. The count is neither too thin nor bloated.
The tool set provides complete coverage for the stated domain of media analysis: images, audio, video, and video with audio. The endpoint status tool adds necessary operational visibility. There are no obvious missing operations for the server's purpose.