- LIVE
- TIMING
- HISTORY
- HUMAN
Short answer
How can I check my Comet ML training runs in chat?
Your AI client uses Comet ML tools to fetch live execution status and timing for your active experiments. You get real-time updates on your runs without leaving your chat interface. This breakdown shows you how it works.
The outcomes
What you get from your agent
The AI turns raw experiment data into direct answers in your chat.
LIVE
Active run status
The AI checks the current state of your training. You'll know if a run is still active or has finished.
TIMING
Execution progress
Your agent pulls timing data for active runs. This lets you track how long training has been running.
HISTORY
Experiment lookups
The AI finds specific experiments from your project history. You can quickly pull details on past runs.
HUMAN
Manual control
The AI only reads data. You still need to use the Comet ML dashboard to stop or edit any runs.
The workflow
What your AI does when the status is requested.
The AI handles the data retrieval from Comet ML so you can stay in your workflow.
Find experiments
The agent looks through your Comet ML projects to find the specific runs you're asking about.
list_experimentsGet run details
Once the agent identifies the right run, it pulls the specific execution data.
get_experimentCheck timing
The agent uses experiment data to report on how long a run has been active.
get_experimentSummarize status
The agent combines the retrieved information to give you a clear status update in chat.
list_experimentsget_experiment
Try it
Copy these to start.
Use these prompts to start checking your experiments.
Starting points
Swap out the project names and experiment IDs for your own specific Comet ML identifiers.
Comet ML Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_C2PxqYqwffQi3V41WE8jlPynFSP048sA8SVR1PFT/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Comet ML capabilities are ready to use.
{
"mcpServers": {
"comet-ml-mcp": {
"url": "https://edge.vinkius.com/vk_preview_C2PxqYqwffQi3V41WE8jlPynFSP048sA8SVR1PFT/mcp"
}
}
}List all experiments in my 'resnet-training' project.
What is the status of experiment 'run_9921_x'?
How long has the active run in my 'transformer-v2' project been running?
Show me the latest experiments from the 'bert-finetune' project.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Comet ML once, then ask.
Connect your account via the link below. Your credentials stay encrypted, and you can immediately start asking your agent about your runs.
Connect Comet ML to your AIFAQ
How this task behaves
- 01
Can the AI stop a training run?
No. The AI can only read and list experiment data. You must stop runs manually in Comet ML.
- 02
Can the AI change my hyperparameters?
No. The tools only allow the AI to fetch existing data, not modify experiment configurations.
- 03
Can the AI delete experiments?
No. The AI has no capability to delete any data from Comet ML.
- 04
Does the AI see all my projects?
The AI can access any project that your authenticated account has permission to view.
- 05
How does the AI get the timing data?
The agent pulls the execution details directly from the specific experiment you've identified.
More questions about Comet ML? The Connector page answers them. See everything the Comet ML Connector can do
Connect Comet ML to Claude, Cursor, ChatGPT & more
