Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Qwen3-VL Video Understanding MCP Server Summarize this product demo video: https://example.com/demo.mp4"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP (Model Context Protocol) server that enables Claude and other AI agents to analyze videos and images using Qwen3-VL-8B-Instruct deployed on Blaxel.
Video Analysis: Analyze videos via URL with custom prompts
Image Analysis: Analyze images via URL
Video Summarization: Generate summaries in different styles
Text Extraction: Extract on-screen text and transcribe speech
Video Q&A: Ask specific questions about video content
H100 GPUs: Fast inference on NVIDIA H100 GPUs via Blaxel
Related MCP server: Vision MCP Server
Claude/Agent → MCP Server → Blaxel API → Qwen3-VL (H100 GPUs)Blaxel Account: Sign up at blaxel.ai
Blaxel CLI: Install the Blaxel CLI
ffmpeg: Required for video frame extraction
Python 3.10+
cat << 'EOF' | blaxel apply -f -
apiVersion: blaxel.ai/v1alpha1
kind: Model
metadata:
name: qwen-qwen3-vl-8b-instruct
displayName: Qwen/Qwen3-VL-8B-Instruct
spec:
enabled: true
policies: []
flavors:
- name: nvidia-h100/x4
type: gpu
runtime:
model: Qwen/Qwen3-VL-8B-Instruct
type: hf_private_endpoint
image: ''
args: []
endpointName: qwenqwen3-vl-8b-instruct-nvidia-h100
organization: adamanz
integrationConnections:
- huggingface-4s2m2h
EOFOr use the provided config:
blaxel apply -f blaxel-model.yamlblaxel auth tokencd qwen-video-blaxel-mcp
pip install -e .Or with uv:
uv pip install -e .cp .env.example .env
# Edit .env with your Blaxel API keyAdd to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"qwen3-video-blaxel": {
"command": "uv",
"args": [
"--directory",
"/path/to/qwen-video-blaxel-mcp",
"run",
"server.py"
],
"env": {
"BLAXEL_API_KEY": "your-blaxel-api-key",
"BLAXEL_MODEL": "qwen-qwen3-vl-8b-instruct"
}
}
}
}The qwen3-video-blaxel tools should now be available.
analyze_videoAnalyze a video with a custom prompt.
analyze_video(
video_url="https://example.com/video.mp4",
question="What happens in this video?",
max_frames=8
)analyze_imageAnalyze an image with a custom prompt.
analyze_image(
image_url="https://example.com/image.jpg",
question="Describe this image"
)summarize_videoGenerate a video summary.
summarize_video(
video_url="https://example.com/video.mp4",
style="detailed" # brief, standard, or detailed
)video_qaAsk specific questions about a video.
video_qa(
video_url="https://example.com/video.mp4",
question="How many people appear?"
)extract_video_textExtract text and transcribe speech.
extract_video_text(
video_url="https://example.com/presentation.mp4"
)check_configurationCheck the Blaxel API configuration.
list_capabilitiesList all server capabilities.
Environment Variable | Description | Default |
| Your Blaxel API key | Required |
| Blaxel API URL |
|
| Model name |
|
ffmpeg: Required for video frame extraction
# macOS
brew install ffmpeg
# Ubuntu/Debian
apt install ffmpegVideo: mp4, webm, mov, avi
Image: jpg, jpeg, png, gif, webp
Feature | Modal | Blaxel |
Model | Qwen2.5-VL-7B | Qwen3-VL-8B |
GPU | A100 | H100 |
Pricing | Pay-per-second | Subscription |
Cold Start | ~30-60s | Faster |
Setup | Deploy code | Apply YAML |
MIT
Analyze an image using Qwen3-VL-8B vision-language model on Blaxel.
The image must be accessible via a public URL.
Examples:
- "What's in this image?"
- "Describe the scene"
- "What text is visible?"
- "Identify any people or objects"
| Name | Required | Description | Default |
|---|---|---|---|
| image_url | Yes | URL of the image to analyze (must be publicly accessible) | |
| question | No | Question or prompt about the image | Describe this image in detail. |
| max_tokens | No | Maximum tokens in response |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It adds useful context beyond the input schema by specifying that the image must be publicly accessible via URL, which is a key constraint. However, it lacks details on rate limits, authentication needs, error handling, or response format, leaving gaps in behavioral understanding.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and front-loaded, starting with the core purpose and key constraint. The examples are concise and relevant, adding practical value without unnecessary elaboration. Every sentence earns its place by enhancing clarity and usability.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity (3 parameters, no output schema, no annotations), the description is partially complete. It covers the purpose and key constraint well but lacks details on behavioral aspects like response format, error cases, or performance characteristics. Without annotations or output schema, more context would improve completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all parameters thoroughly. The description does not add significant meaning beyond the schema, such as explaining parameter interactions or providing additional examples. The baseline score of 3 is appropriate since the schema handles the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the specific action ('analyze an image'), the resource ('image'), and the method ('using Qwen3-VL-8B vision-language model on Blaxel'). It distinguishes from siblings like 'analyze_video' and 'video_qa' by focusing exclusively on image analysis, not video processing or configuration tasks.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context for when to use this tool ('analyze an image') and includes practical examples of typical use cases. However, it does not explicitly state when NOT to use it or name specific alternatives among siblings, such as 'analyze_video' for video content or 'extract_video_text' for text extraction from videos.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Analyze a video using Qwen3-VL-8B vision-language model on Blaxel.
The video must be accessible via a public URL. The model will:
1. Download the video
2. Extract key frames (up to max_frames)
3. Analyze the frames with your question
Examples:
- "What happens in this video?"
- "Summarize the main events"
- "What products are shown?"
- "Describe the people and their actions"
| Name | Required | Description | Default |
|---|---|---|---|
| video_url | Yes | URL of the video to analyze (must be publicly accessible) | |
| question | No | Question or prompt about the video | Describe what happens in this video in detail. |
| max_frames | No | Maximum number of frames to extract (1-16) | |
| max_tokens | No | Maximum tokens in response |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It adds useful context about the process (download video, extract frames, analyze frames) and constraints (video must be publicly accessible). However, it lacks details on rate limits, authentication needs, error handling, or response format, which are important for a tool with potential computational and network implications.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and front-loaded, starting with the core purpose. Each sentence earns its place: the first states the action, the second specifies a key constraint, the third outlines the process steps, and the examples illustrate usage without redundancy. It's efficient and well-structured for quick understanding.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of video analysis with 4 parameters, no annotations, and no output schema, the description is mostly complete. It covers the purpose, process, constraints, and usage examples, but lacks details on behavioral aspects like performance, limitations, or output format, which would be helpful for an AI agent to manage expectations and errors.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all parameters thoroughly. The description adds minimal value beyond the schema by implying the purpose of parameters through examples (e.g., question examples relate to the 'question' parameter), but it doesn't provide additional syntax, format, or usage details. This meets the baseline of 3 when schema coverage is high.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the specific action ('analyze a video using Qwen3-VL-8B vision-language model on Blaxel') and distinguishes it from siblings by focusing on comprehensive video analysis rather than text extraction (extract_video_text), simple Q&A (video_qa), or summarization (summarize_video). It specifies the resource (video) and method (vision-language model).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context for when to use this tool (analyzing video content with questions) through examples like 'What happens in this video?' and 'Summarize the main events.' However, it doesn't explicitly state when not to use it or name specific alternatives among siblings, such as when to choose summarize_video instead for pure summarization without custom questions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Check the Blaxel API configuration.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden. It states the tool 'checks' configuration, implying a read-only operation, but doesn't disclose behavioral traits like what happens if configuration is invalid, whether it returns status details, or if it has side effects. This leaves significant gaps in understanding the tool's behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence with no wasted words. It's front-loaded and appropriately sized for a simple tool, making it easy to parse quickly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no annotations and no output schema, the description is incomplete. It doesn't explain what the check entails, what output to expect, or how results are formatted. For a configuration-checking tool, this leaves the agent with insufficient context to use it effectively.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters with 100% schema description coverage, so no parameter information is needed. The description doesn't add param details, which is appropriate, earning a baseline score of 4 for not introducing confusion or redundancy.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states the tool checks Blaxel API configuration, which is a clear purpose but lacks specificity about what aspects are checked or what 'configuration' entails. It doesn't differentiate from sibling tools, which are all media analysis tools, making this tool's purpose distinct by default but not explicitly contrasted.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus alternatives. The description doesn't mention prerequisites, context, or exclusions, leaving the agent to infer usage based on the tool name alone.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Extract text and transcribe speech from a video.
| Name | Required | Description | Default |
|---|---|---|---|
| video_url | Yes | URL of the video |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It mentions what the tool does but doesn't cover important aspects like performance characteristics (e.g., processing time, accuracy), error handling, or output format. This leaves significant gaps for an AI agent to understand how to use it effectively.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise and front-loaded, consisting of a single sentence that directly states the tool's purpose without any unnecessary words. Every part of the sentence earns its place by clearly communicating the core functionality.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of video processing and the lack of annotations and output schema, the description is insufficiently complete. It doesn't explain what the output looks like (e.g., text format, timestamps, confidence scores) or address potential limitations (e.g., supported video formats, language support), leaving the AI agent with incomplete context for proper tool invocation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 100% description coverage, with the single parameter 'video_url' clearly documented. The description doesn't add any parameter-specific information beyond what's in the schema, but since schema coverage is high, the baseline score of 3 is appropriate as the schema adequately handles parameter documentation.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with specific verbs ('extract text' and 'transcribe speech') and resource ('from a video'), making it easy to understand what it does. However, it doesn't differentiate from sibling tools like 'analyze_video' or 'summarize_video', which might have overlapping functionality.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. With siblings like 'analyze_video', 'summarize_video', and 'video_qa' available, there's no indication of when this specific text extraction tool is preferred, leaving usage context unclear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all server capabilities.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It states a read operation ('List'), implying it's likely safe and non-destructive, but doesn't disclose behavioral traits like rate limits, authentication needs, response format, or potential side effects. This is inadequate for a tool with zero annotation coverage.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence with zero waste. It's appropriately sized for a simple tool and front-loaded with the core action, making it easy to parse quickly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the lack of annotations and output schema, the description is incomplete. It doesn't explain what 'capabilities' entail, the return format, or any behavioral context, leaving significant gaps for the agent to understand the tool's full scope and usage.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters, and schema description coverage is 100%, so the schema fully documents the lack of inputs. The description doesn't need to add parameter details, earning a baseline score of 4 for not introducing confusion or redundancy.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('List') and resource ('all server capabilities'), making the purpose understandable. However, it doesn't differentiate from sibling tools (like 'check_configuration'), which might have overlapping functionality, preventing a perfect score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives, such as 'check_configuration' or other siblings. It lacks context about prerequisites, timing, or exclusions, leaving the agent with minimal usage direction.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Generate a summary of a video.
Styles:
- brief: 1-2 sentence overview
- standard: 1-2 paragraph summary
- detailed: Comprehensive analysis
| Name | Required | Description | Default |
|---|---|---|---|
| video_url | Yes | URL of the video to summarize | |
| style | No | Summary style: 'brief', 'standard', or 'detailed' | standard |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It mentions the tool generates summaries with style options but doesn't describe what the output looks like (e.g., text format, length limits), whether it requires authentication, rate limits, or how it handles errors. For a tool with no annotations and no output schema, this leaves significant behavioral gaps.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and front-loaded: the first sentence states the core purpose, followed by a bulleted list of style options that efficiently explains variations. Every sentence earns its place with no redundant or vague language, making it easy to scan and understand quickly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool has no annotations, no output schema, and 2 parameters with full schema coverage, the description is incomplete. It lacks information on output format, error handling, authentication needs, and how it differs from sibling tools. For a summarization tool with no structured behavioral data, the description should provide more context to guide effective use.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents both parameters ('video_url' and 'style') with descriptions and default values. The description adds minimal value by listing style options ('brief', 'standard', 'detailed'), which are already in the schema's description for 'style'. It doesn't provide additional context like URL format requirements or style-specific details beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Generate a summary of a video' with a specific verb ('Generate') and resource ('video'). It distinguishes from siblings like 'analyze_video' or 'extract_video_text' by focusing on summarization rather than analysis or text extraction. However, it doesn't explicitly differentiate from all siblings (e.g., 'video_qa' might also involve summarization).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It doesn't mention when to choose 'summarize_video' over 'analyze_video' or 'video_qa', nor does it specify prerequisites or exclusions. The style options imply usage contexts but don't offer explicit alternatives or when-not-to-use advice.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Ask a specific question about a video's content.
Examples:
- "How many people appear?"
- "What color is the car?"
- "What is being demonstrated?"
| Name | Required | Description | Default |
|---|---|---|---|
| video_url | Yes | URL of the video | |
| question | Yes | Your specific question about the video |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It mentions the tool asks questions about video content but doesn't describe how it processes videos (e.g., via AI models), potential limitations (e.g., accuracy, video length), or response format. This is a significant gap for a tool with no annotation coverage.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is front-loaded with a clear purpose statement, followed by concise examples that illustrate usage without unnecessary details. Every sentence earns its place by reinforcing the tool's function, making it efficient and well-structured for quick understanding.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of video analysis and the lack of annotations and output schema, the description is incomplete. It doesn't explain behavioral aspects like how videos are processed, what types of questions are supported, or what the return values might be. This leaves the agent with insufficient context for reliable tool invocation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents both parameters ('video_url' and 'question') with clear descriptions. The description adds minimal value beyond the schema by implying the question should be specific and providing examples, but it doesn't elaborate on parameter constraints or usage nuances. Baseline 3 is appropriate given the high schema coverage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose as 'Ask a specific question about a video's content,' which includes a verb ('Ask') and resource ('video's content'). It distinguishes from siblings like 'summarize_video' or 'analyze_video' by focusing on Q&A rather than summarization or general analysis. However, it could be more specific about the type of questions supported beyond the examples.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage through examples (e.g., 'How many people appear?'), suggesting it's for specific, factual questions about visual content. It doesn't explicitly state when to use this tool versus alternatives like 'analyze_video' or 'extract_video_text,' nor does it provide exclusions or prerequisites, leaving some ambiguity for the agent.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0analyze_imageanalyze_videocheck_configurationextract_video_textlist_capabilitiessummarize_videovideo_qaScored across 7 tools
The tools have overlapping purposes that could cause confusion. analyze_video, summarize_video, and video_qa all process videos with similar capabilities, while analyze_image stands alone for images. The descriptions help differentiate use cases, but an agent might struggle to choose between summarize_video and video_qa for general video queries.
Most tools follow a consistent verb_noun pattern (analyze_image, analyze_video, extract_video_text, summarize_video, video_qa). However, check_configuration and list_capabilities deviate slightly with different verb styles, though they remain readable and understandable.
With 7 tools, the count is well-scoped for a video understanding server. Each tool has a distinct role, covering image analysis, video analysis, configuration checks, text extraction, and capability listing, making the set comprehensive without being overwhelming.
The tool surface covers core video and image analysis tasks well, including configuration and capability checks. Minor gaps exist, such as no explicit tool for editing or modifying media, but agents can work around this with the provided analysis and summarization tools for typical use cases.
Video, audio, and image processing for AI agents: convert, transcribe, upscale - 150+ operations.
Multimodal video analysis MCP — transcription, vision, and OCR for any video URL.
15 media & data tools for AI agents: search, transcribe, subtitles, voiceover, translate & more.
PDF, image, video, OCR, screenshot, SQL, QR and text tools for agents. No API key, no signup.
