Contamination MapTime series / Model
Granite-TTM-R3
- Organisation
- IBM Research
- Family
- Tiny Time Mixers (TTM) · version R3
- Released
- 2026-05-21 (HF repo creation (createdAt) of ibm-granite/granite-timeseries-ttm-r3 (weights uploaded to main 2026-07-02), https://huggingface.co/api/models/ibm-granite/granite-timeseries-ttm-r3)
- Checkpoints
- ibm-granite/granite-timeseries-ttm-r3, ibm-granite/granite-timeseries-ttm-r3@1024-48-dec-512-lite-r3, ibm-granite/granite-timeseries-ttm-r3@1024-48-dec-512-r3, ibm-granite/granite-timeseries-ttm-r3@1024-720-lite-r3, ibm-granite/granite-timeseries-ttm-r3@1024-720-r3, ibm-granite/granite-timeseries-ttm-r3@1024-96-lite-r3, ibm-granite/granite-timeseries-ttm-r3@1024-96-r3, ibm-granite/granite-timeseries-ttm-r3@1536-720-lite-r3, ibm-granite/granite-timeseries-ttm-r3@1536-720-r3, ibm-granite/granite-timeseries-ttm-r3@1536-96-lite-r3, ibm-granite/granite-timeseries-ttm-r3@1536-96-r3, ibm-granite/granite-timeseries-ttm-r3@156-16-dec-52-lite-r3, ibm-granite/granite-timeseries-ttm-r3@156-16-dec-52-r3, ibm-granite/granite-timeseries-ttm-r3@180-60-dec-180-lite-r3, ibm-granite/granite-timeseries-ttm-r3@180-60-dec-180-r3, ibm-granite/granite-timeseries-ttm-r3@2048-720-lite-r3, ibm-granite/granite-timeseries-ttm-r3@2048-720-r3, ibm-granite/granite-timeseries-ttm-r3@2048-96-lite-r3, ibm-granite/granite-timeseries-ttm-r3@2048-96-r3, ibm-granite/granite-timeseries-ttm-r3@2560-720-lite-r3, ibm-granite/granite-timeseries-ttm-r3@2560-720-r3, ibm-granite/granite-timeseries-ttm-r3@2560-96-lite-r3, ibm-granite/granite-timeseries-ttm-r3@2560-96-r3, ibm-granite/granite-timeseries-ttm-r3@3072-720-lite-r3, ibm-granite/granite-timeseries-ttm-r3@3072-720-r3, ibm-granite/granite-timeseries-ttm-r3@3072-96-lite-r3, ibm-granite/granite-timeseries-ttm-r3@3072-96-r3, ibm-granite/granite-timeseries-ttm-r3@360-60-dec-360-lite-r3, ibm-granite/granite-timeseries-ttm-r3@360-60-dec-360-r3, ibm-granite/granite-timeseries-ttm-r3@512-30-dec-90-lite-r3, ibm-granite/granite-timeseries-ttm-r3@512-30-dec-90-r3, ibm-granite/granite-timeseries-ttm-r3@512-336-dec-512-lite-r3, ibm-granite/granite-timeseries-ttm-r3@512-336-dec-512-r3, ibm-granite/granite-timeseries-ttm-r3@512-48-dec-512-lite-r3, ibm-granite/granite-timeseries-ttm-r3@512-48-dec-512-r3, ibm-granite/granite-timeseries-ttm-r3@512-96-dec-512-lite-r3, ibm-granite/granite-timeseries-ttm-r3@512-96-dec-512-r3, ibm-granite/granite-timeseries-ttm-r3@52-16-dec-52-lite-r3, ibm-granite/granite-timeseries-ttm-r3@52-16-dec-52-r3, ibm-granite/granite-timeseries-ttm-r3@768-48-dec-512-lite-r3, ibm-granite/granite-timeseries-ttm-r3@768-48-dec-512-r3, ibm-granite/granite-timeseries-ttm-r3@90-30-dec-90-lite-r3, ibm-granite/granite-timeseries-ttm-r3@90-30-dec-90-r3
- Weights
- Open · apache-2.0
- Links
- Model cardCode
Training data disclosure: Partial list
Level determined for 61% of benchmark cells.
- TestTest data seen0
- SeenSeen, split unclear1
- TrainTraining data seen0
- Related data seen1
- NoNot seen28
- ?Unknown19
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
“Custom synthesized data based on [KernelSynth](https://github.com/amazon-science/chronos-forecasting/blob/main/scripts/kernel-synth.py).”
Training corpora
-
GIFT-Eval Pretrain (GiftEvalPretrain)
· 86 sources (complete list) · pre-training
· homepage
Evidence that Granite-TTM-R3 uses it
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
huggingface.co · model card · Training Data section · retrieved 2026-10-15 Contents of this corpus
-
BDG-2 Panther · split: unknown
HF folder(s): bdg-2_panther.
Evidence
“BDG-2 Panther”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
BDG-2 Fox · split: unknown
HF folder(s): bdg-2_fox.
Evidence
“BDG-2 Fox BDG-2 Rat”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
BDG-2 Rat · split: unknown
HF folder(s): bdg-2_rat.
Evidence
“BDG-2 Rat BDG-2 Bear”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
BDG-2 Bear · split: unknown
HF folder(s): bdg-2_bear.
Evidence
“BDG-2 Bear”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Low Carbon London · split: unknown
HF folder(s): lcl.
Evidence
“Low Carbon London”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
SMART · split: unknown
HF folder(s): smart.
Evidence
“SMART IDEAL”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
IDEAL · split: unknown
HF folder(s): ideal.
Evidence
“IDEAL Sceaux”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Sceaux · split: unknown
HF folder(s): sceaux.
Evidence
“Sceaux Borealis”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Borealis · split: unknown
HF folder(s): borealis.
Evidence
“Borealis Buildings900K”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Buildings900K · split: unknown
HF folder(s): buildings_900k.
Evidence
“Buildings900K”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
CMIP6 · split: unknown
HF folder(s): cmip6_1850 ... cmip6_2010 (33 folders, every 5 years).
Evidence
“CMIP6 ERA5”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
ERA5 · split: unknown
HF folder(s): era5_1989 ... era5_2018 (30 yearly folders).
Evidence
“ERA5 Azure VM Traces 2017”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Azure VM Traces 2017 · split: unknown
HF folder(s): azure_vm_traces_2017.
Evidence
“Azure VM Traces 2017”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Borg Cluster Data 2011 · split: unknown
HF folder(s): borg_cluster_data_2011.
Evidence
“Borg Cluster Data 2011”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Alibaba Cluster Trace 2018 · split: unknown
HF folder(s): alibaba_cluster_trace_2018.
Evidence
“Alibaba Cluster Trace 2018”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Taxi · split: unknown
HF folder(s): taxi_30min.
Evidence
“Taxi Uber TLC Daily”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Uber TLC Daily · split: unknown
HF folder(s): uber_tlc_daily.
Evidence
“Uber TLC Daily”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Uber TLC Hourly · split: unknown
HF folder(s): uber_tlc_hourly.
Evidence
“Uber TLC Hourly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Wiki-Rolling · split: unknown
HF folder(s): wiki-rolling_nips.
Evidence
“Wiki-Rolling”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
M5 · split: unknown
HF folder(s): m5.
Evidence
“M5 LargeST”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
LargeST (PeMS California) · split: unknown
HF folder(s): largest_2017 ... largest_2021 (5 yearly folders).
Evidence
“LargeST PEMS03”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
PEMS03 · split: unknown
HF folder(s): PEMS03.
Evidence
“PEMS03 PEMS04”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
PEMS04 · split: unknown
HF folder(s): PEMS04.
Evidence
“PEMS04 PEMS07”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
PEMS07 · split: unknown
HF folder(s): PEMS07.
Evidence
“PEMS07 PEMS08”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
PEMS08 · split: unknown
HF folder(s): PEMS08.
Evidence
“PEMS08 PEMS Bay”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
PEMS Bay · split: unknown
HF folder(s): PEMS_BAY.
Evidence
“PEMS Bay Los-Loop”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Los-Loop · split: unknown
HF folder(s): LOS_LOOP.
Evidence
“Los-Loop Beijing Subway”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Beijing Subway · split: unknown
HF folder(s): BEIJING_SUBWAY_30MIN.
Evidence
“Beijing Subway”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
SHMetro · split: unknown
HF folder(s): SHMETRO.
Evidence
“SHMetro HZMetro”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
HZMetro · split: unknown
HF folder(s): HZMETRO.
Evidence
“HZMetro Q-Traffic”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Q-Traffic · split: unknown
HF folder(s): Q-TRAFFIC.
Evidence
“Q-Traffic Subseasonal”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Subseasonal · split: unknown
HF folder(s): subseasonal.
Evidence
“Subseasonal”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Subseasonal Precipitation · split: unknown
HF folder(s): subseasonal_precip.
Evidence
“Subseasonal Precipitation”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Covid19 Energy · split: unknown
HF folder(s): covid19_energy.
Evidence
“Covid19 Energy”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
GEF12 · split: unknown
HF folder(s): gfc12_load.
Evidence
“GEF12 GEF14”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
GEF14 · split: unknown
HF folder(s): gfc14_load.
Evidence
“GEF14 GEF17”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
GEF17 · split: unknown
HF folder(s): gfc17_load.
Evidence
“GEF17 PDB Spanish”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
PDB · split: unknown
HF folder(s): pdb.
Evidence
“PDB Spanish”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Spanish · split: unknown
HF folder(s): spain.
Evidence
“Spanish BDG-2 Hog”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
BDG-2 Hog · split: unknown
HF folder(s): hog.
Evidence
“BDG-2 Hog BDG-2 Bull”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
BDG-2 Bull · split: unknown
HF folder(s): bull.
Evidence
“BDG-2 Bull”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
BDG-2 Cockatoo · split: unknown
HF folder(s): cockatoo.
Evidence
“BDG-2 Cockatoo”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
ELF · split: unknown
HF folder(s): elf.
Evidence
“ELF London Smart Meters”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
London Smart Meters · split: unknown
HF folder(s): london_smart_meters_with_missing.
Evidence
“London Smart Meters”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Wind Farms · split: unknown
HF folder(s): wind_farms_with_missing.
Evidence
“Wind Farms”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Wind Power · split: unknown
HF folder(s): wind_power.
Evidence
“Wind Power”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Solar Power · split: unknown
HF folder(s): solar_power. AEMO 4-second solar power (Monash), not the LSTNet Alabama solar dataset in the GIFT-Eval test set.
Evidence
“Solar Power”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Oikolab Weather · split: unknown
HF folder(s): oikolab_weather.
Evidence
“Oikolab Weather”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Elecdemand · split: unknown
HF folder(s): elecdemand.
Evidence
“Elecdemand”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Covid Mobility · split: unknown
HF folder(s): covid_mobility.
Evidence
“Covid Mobility”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Kaggle Web Traffic Weekly · split: unknown
HF folder(s): kaggle_web_traffic_weekly.
Evidence
“Kaggle Web Traffic Weekly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Extended Web Traffic · split: unknown
HF folder(s): extended_web_traffic_with_missing.
Evidence
“Extended Web Traffic”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
M1 Yearly · split: unknown
HF folder(s): m1_yearly.
Evidence
“M1 Yearly M1 Quarterly”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
M1 Quarterly · split: unknown
HF folder(s): m1_quarterly.
Evidence
“M1 Quarterly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
M1 Monthly · split: unknown
HF folder(s): m1_monthly.
Evidence
“M1 Monthly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
M3 Yearly · split: unknown
HF folder(s): monash_m3_yearly.
Evidence
“M3 Yearly M3 Quarterly”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
M3 Quarterly · split: unknown
HF folder(s): monash_m3_quarterly.
Evidence
“M3 Quarterly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
M3 Monthly · split: unknown
HF folder(s): monash_m3_monthly.
Evidence
“M3 Monthly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
M3 Other · split: unknown
HF folder(s): monash_m3_other.
Evidence
“M3 Other NN5 Daily”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
NN5 Daily · split: unknown
HF folder(s): nn5_daily_with_missing.
Evidence
“NN5 Daily NN5 Weekly”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
NN5 Weekly · split: unknown
HF folder(s): nn5_weekly.
Evidence
“NN5 Weekly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Tourism Yearly · split: unknown
HF folder(s): tourism_yearly.
Evidence
“Tourism Yearly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Tourism Quarterly · split: unknown
HF folder(s): tourism_quarterly.
Evidence
“Tourism Quarterly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Tourism Monthly · split: unknown
HF folder(s): tourism_monthly.
Evidence
“Tourism Monthly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
CIF 2016 · split: unknown
HF folder(s): cif_2016_12, cif_2016_6.
Evidence
“CIF 2016 Traffic Weekly”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Traffic Weekly (Monash) · split: unknown
HF folder(s): traffic_weekly. Monash weekly aggregation of the 862-sensor traffic dataset (registry id monash-traffic-weekly).
Evidence
“Traffic Weekly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Traffic (PeMS San Francisco Bay, 862 sensors) · split: unknown
HF folder(s): traffic_hourly. Monash 'traffic_hourly' = the 862-sensor San Francisco traffic dataset of LSTNet, i.e. the long-horizon Traffic dataset (registry id traffic-862).
Evidence
“Traffic Hourly”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Australian Electricity Demand · split: unknown
HF folder(s): australian_electricity_demand.
Evidence
“Australian Electricity Demand”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Rideshare · split: unknown
HF folder(s): rideshare_with_missing.
Evidence
“Rideshare Sunspot”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Sunspot · split: unknown
HF folder(s): sunspot_with_missing.
Evidence
“Sunspot Vehicle Trips”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Vehicle Trips · split: unknown
HF folder(s): vehicle_trips_with_missing.
Evidence
“Vehicle Trips”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Weather · split: unknown
HF folder(s): weather. Monash 'weather' (Australian BOM, 3,010 daily series; Table 14 source column: Monash), NOT the Jena weather data used in GIFT-Eval and long-horizon.
Evidence
“Weather FRED MD”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
FRED MD · split: unknown
HF folder(s): fred_md.
Evidence
“FRED MD Pedestrian Counts”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Pedestrian Counts · split: unknown
HF folder(s): pedestrian_counts.
Evidence
“Pedestrian Counts”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Bitcoin · split: unknown
HF folder(s): bitcoin_with_missing.
Evidence
“Bitcoin KDD Cup 2022”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
KDD Cup 2022 · split: unknown
HF folder(s): kdd2022. Not the KDD Cup 2018 air-quality dataset that is in the GIFT-Eval test set.
Evidence
“KDD Cup 2022”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
GoDaddy · split: unknown
HF folder(s): godaddy.
Evidence
“GoDaddy Favorita Sales”
arxiv.org · paper · Table 14 (dataset column; following row names included to reach the minimum quote length) · retrieved 2026-10-15 -
Favorita Sales · split: unknown
HF folder(s): favorita_sales.
Evidence
“Favorita Sales”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Favorita Transactions · split: unknown
HF folder(s): favorita_transactions.
Evidence
“Favorita Transactions”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
China Air Quality · split: unknown
HF folder(s): china_air_quality.
Evidence
“China Air Quality”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Beijing Air Quality · split: unknown
HF folder(s): beijing_air_quality.
Evidence
“Beijing Air Quality”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Residential Load Power · split: unknown
HF folder(s): residential_load_power.
Evidence
“Residential Load Power”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Residential PV Power · split: unknown
HF folder(s): residential_pv_power.
Evidence
“Residential PV Power”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
CDC FluView ILINet (LOTSA) · split: unknown
HF folder(s): cdc_fluview_ilinet. Same source (CDC FluView) as the long-horizon ILI dataset (registry id cdc-fluview-ilinet).
Evidence
“CDC Fluview ILINet”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
CDC Fluview WHO NREVSS · split: unknown
HF folder(s): cdc_fluview_who_nrevss.
Evidence
“CDC Fluview WHO NREVSS”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15 -
Project Tycho · split: unknown
HF folder(s): project_tycho.
Evidence
“Project Tycho”
arxiv.org · paper · Table 14 (dataset column) · retrieved 2026-10-15
-
BDG-2 Panther · split: unknown
Other declared training data
-
GIFT-Eval Train (not yet matched to a dataset) · split: train of gift-eval
Model card links Salesforce/GiftEval (the GIFT-Eval benchmark dataset) and says only non-leaking historical context was used.
Evidence
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
huggingface.co · model card · Training Data section · retrieved 2026-10-15
Declared exclusions
-
Benchmark GIFT-Eval · test portion excluded
Evidence
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
huggingface.co · model card · Training Data section · retrieved 2026-10-15
Benchmark cells
GIFT-Eval
Not seen Jena Weather (GIFT-Eval) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen BizITObs - Application · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen BizITObs - Service · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen BizITObs - L2C · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Bitbrains - Fast Storage · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Bitbrains - rnd · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Restaurant (Recruit Restaurant Visitor Forecasting) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen ETT1 (GIFT-Eval) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen ETT2 (GIFT-Eval) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Loop Seattle (LibCity) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen SZ-Taxi (LibCity) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen M_DENSE (LibCity) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Solar (LSTNet, 137 Alabama PV plants) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Hierarchical Sales (Mancuso et al., 2020) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen M4 Yearly · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen M4 Quarterly · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen M4 Monthly · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen M4 Weekly · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen M4 Daily · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen M4 Hourly · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Hospital · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen COVID Deaths · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen US Births · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Saugeen River Flow · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Temperature Rain · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen KDD Cup 2018 · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen Car Parts · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Not seen ElectricityLoadDiagrams20112014 (UCI) · the authors state its test portion was excluded from training
Evidence (1)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
Long-horizon forecasting (ETT, Electricity, Traffic, Weather, Exchange, ILI)
Unknown ETTh1 · training data not disclosed in enough detail
Unknown ETTh2 · training data not disclosed in enough detail
Unknown ETTm1 · training data not disclosed in enough detail
Unknown ETTm2 · training data not disclosed in enough detail
Unknown Electricity (ECL, 321 clients) · training data not disclosed in enough detail
Seen, split unclear Traffic (PeMS San Francisco Bay, 862 sensors) · the dataset was used; the source does not say which part (via GIFT-Eval Pretrain (GiftEvalPretrain))
Evidence (2)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
“Traffic Hourly”
Unknown Weather (Jena, 2020; Autoformer) · training data not disclosed in enough detail
Unknown Exchange-Rate · training data not disclosed in enough detail
ILI (CDC influenza-like illness, national) · related dataset CDC FluView ILINet (LOTSA) was used (via GIFT-Eval Pretrain (GiftEvalPretrain))
Evidence (5)
“Select Datasets from [GiftEvalPretrain](https://huggingface.co/datasets/Salesforce/GiftEvalPretrain) and [GIFT-Eval Train](https://huggingface.co/datasets/Salesforce/GiftEval), using non-leaking historical context.”
“CDC Fluview ILINet”
“includes the weekly recorded influenza-like illness (ILI) patients data from Centers for Disease Control and Prevention of the United States between 2002 and 2021”
“https://gis.cdc.gov/grasp/fluview/fluportaldashboard.html”
“CDC Fluview ILINet CDC Fluview WHO NREVSS”
Time-series anomaly detection
Unknown TSB-AD-U (TSB-AD univariate) · training data not disclosed in enough detail
Unknown TSB-AD-M (TSB-AD multivariate) · training data not disclosed in enough detail
Unknown UCR Time Series Anomaly Archive (Hexagon ML/UCR, KDD Cup 2021) · training data not disclosed in enough detail
Unknown Server Machine Dataset (SMD) · training data not disclosed in enough detail
Unknown Mars Science Laboratory (MSL) telemetry anomaly dataset · training data not disclosed in enough detail
Unknown Soil Moisture Active Passive (SMAP) telemetry anomaly dataset · training data not disclosed in enough detail
Unknown Pooled Server Metrics (PSM) · training data not disclosed in enough detail
Unknown Secure Water Treatment (SWaT) · training data not disclosed in enough detail
Unknown Yahoo S5 Labeled Anomaly Detection Dataset (Yahoo Webscope S5) · training data not disclosed in enough detail
Unknown Numenta Anomaly Benchmark (NAB) · training data not disclosed in enough detail
Time-series classification
Unknown UCR Time Series Classification Archive (2018) · training data not disclosed in enough detail
Unknown UEA Multivariate Time Series Classification Archive (2018) · training data not disclosed in enough detail
Notes
Apache-2.0 Granite release. Card says 'Select Datasets' from GiftEvalPretrain and GIFT-Eval Train (non-leaking historical context) plus KernelSynth synthetic data; the selection is not listed. Reported GIFT-Eval scores differ from ibm-research/ttm-r3 (e.g. pre-trained MASE 0.727 vs 0.724), so weights are likely different; whether the training data differ is not stated. No R3 paper yet.
Sources
- Granite-TTM-R3 model card · model card · retrieved 2026-10-15