The official implementation of the W&B Models and Weave MCP server.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add weights-biases -- uv pip install git+https://github.com/wandb/wandb-mcp-server~/.codex/config.toml
[mcp_servers.weights-biases]
command = "uv"
args = ["pip", "install", "git+https://github.com/wandb/wandb-mcp-server"]
# codex mcp add weights-biases -- uv pip install git+https://github.com/wandb/wandb-mcp-server~/.cursor/mcp.json
{
"mcpServers": {
"weights-biases": {
"command": "uv",
"args": [
"pip",
"install",
"git+https://github.com/wandb/wandb-mcp-server"
]
}
}
}.vscode/mcp.json
{
"servers": {
"weights-biases": {
"type": "stdio",
"command": "uv",
"args": [
"pip",
"install",
"git+https://github.com/wandb/wandb-mcp-server"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"weights-biases": {
"command": "uv",
"args": [
"pip",
"install",
"git+https://github.com/wandb/wandb-mcp-server"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"weights-biases": {
"command": "uv",
"args": [
"pip",
"install",
"git+https://github.com/wandb/wandb-mcp-server"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"weights-biases": {
"command": "uv",
"args": [
"pip",
"install",
"git+https://github.com/wandb/wandb-mcp-server"
]
}
}
}