time series
Granite TinyTimeMixer (TTM) r2
IBM tiny forecasting model for CPU-only machines; vendor claims to beat larger models, but GIFT-Eval ranks TTM-R2 140 of 143 with test leakage flagged.
Levelmodel
Classtime series
Parameters805K
Contextnot in the notes
Released2024
Opennessopen weights permissive
Runs onCPU server, laptop
Statusactive
Benchmark Results
independent
GIFT-Evaltime series
pretrained, test leakage Yesindependentsaw test data
1.02
00.71.4
Score in MASE (geo-mean, normalised to Seasonal Naive). Lower is better.
Leaderboard rebuilt from the published raw results, 143 entries covering all 97 configurations
pretrained, test leakage Yes; rank is average CRPS rankindependentsaw test data
0.873
00.51
Score in CRPS (geo-mean, normalised to Seasonal Naive). Lower is better.
rank 140 of 143, Leaderboard rebuilt from the published raw results, 143 entries covering all 97 configurations
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).
Lineage
successor
Granite TTM-R1 is the parent (successor)high confidencemodel cardGranite TinyTimeMixer (TTM) r2.1 is a child (successor)high confidencemodel cardWhere this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 4 models. Confidence and evidence are the notes' own.
Industry Use Cases
R&D Topics
Identity
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.
Architecture and Size
- Architecture
- other TinyTimeMixer (MLP-mixer family)
- Total parameters
- 805K sources
- Active parameters
- –
- Checkpoint parameters
- –
- Note
- about 805k to 1M+
- Context length
- not in the notes
- Input
- time series
- Output
- time series
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).
Openness and Access
- Openness
- open weights permissive
- Licence
- Apache License 2.0 sources
- Access routes
- open weights download, API
Where this comes from
The licence file governs over registry tags and aggregator labels.
Deployment
- Runs on
- CPU server, laptop
- Footprint
- CPU-only machines and laptops; watsonx.ai forecast API.
- Runtimes
- not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.
Indonesia
- Indonesian support
- not stated
- Regional languages
- not in the notes
- Jakarta hosting
- unknown
- Languages
- not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Links
DeveloperIBMbig tech, USA
IBMbig tech, USAModelGranite TinyTimeMixer (TTM) r2IBM, time series
Granite TinyTimeMixer (TTM) r2IBM, time seriesR&D TopicsDeep Learning Architectures for Time-Series Forecasting and ClassificationTime-Series Foundation Models
Who Runs Itnone recordedDeployments of these use cases; none names this model.
Models 2
Organisations 1
Sources
- ibm-granite/granite-timeseries-ttm-r2 , Hugging Face huggingface.co https://huggingface.co/ibm-granite/granite-timeseries-ttm-r2
- GIFT Eval - a Hugging Face Space by Salesforce huggingface.co https://huggingface.co/spaces/Salesforce/GIFT-Eval
- Granite Time Series | IBM Granite ibm.com https://www.ibm.com/ae-ar/granite/docs/models/time-series/
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Granite TinyTimeMixer (TTM) r2", https://aurion.semai.id/models/ibm-granite-ttm-r2/, records of October 2026.