Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
TIMESFM_MODEL_IDNoHugging Face model id or local model path.google/timesfm-2.5-200m-pytorch
TIMESFM_BATCH_SIZENoTimesFM per_core_batch_size; raise only after memory is stable.64
TIMESFM_MAX_CONTEXTNoMaximum context points used by the compiled model.1024
TIMESFM_MAX_HORIZONNoMaximum forecast horizon.256
TIMESFM_TORCH_COMPILENoEnable PyTorch compile when loading the model.false
TIMESFM_RETURN_BACKCASTNoInternal default; XReg requests force this to true.false
TIMESFM_NORMALIZE_INPUTSNoNormalize each input series before forecasting.true
TIMESFM_INFER_IS_POSITIVENoClamp positive-only series to nonnegative outputs.true
TIMESFM_FIX_QUANTILE_CROSSINGNoEnforce monotonic quantiles.true
TIMESFM_FORCE_FLIP_INVARIANCENoEnforce sign symmetry.true
TIMESFM_CONTINUOUS_QUANTILE_HEADNoUse continuous quantile head for better bands.true

Capabilities

Server capabilities have not been inspected yet.

Tools

Functions exposed to the LLM to take actions

NameDescription

No tools

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/chokukil/timesfm-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server