Toto-2.0-4m
onlineDatadog/Toto-2.0-4m4M params | 512 context | $0.00025 per forecast | Apache-2.0
Toto-2.0-4m is the smallest checkpoint in Datadog's Toto 2.0 family and the entry rung of the family's size ladder. At 4M parameters it brings the new u-muP-scaled architecture, alternating time/variate attention, and direct quantile forecasts down to a footprint suited to quick smoke tests, local-style evaluation, and low-cost observability experiments.
Like every Toto 2.0 checkpoint it is a u-muP-scaled decoder-only transformer with alternating time/variate attention and a quantile output head, so even at this size it produces native probabilistic, multivariate forecasts rather than bare point estimates. Toto 2.0 continues Datadog's observability-first pretraining line for sparse, high-dimensional telemetry, and Datadog positions the release as the current recommended zero-shot generation for BOOM-style infrastructure metrics.
On TSFM.ai, choose Toto-2.0-4m when you are validating an integration, comparing against a baseline, or running cost-sensitive experiments where raw accuracy is secondary. When you need more capacity, move up the ladder to Datadog/Toto-2.0-22m for low-latency production, to the recommended Datadog/Toto-2.0-313m for the main accuracy tier, or to Datadog/Toto-2.0-1B for the highest-value multivariate telemetry workloads.
Model Classification
Family
Toto
Type
time series foundation model
Pretrained time-series model exposed on TSFM.ai for zero-shot or few-shot forecasting workloads.
Resources
Training Data
Toto 2.0 continues Datadog's observability-first pretraining line for sparse, high-dimensional telemetry. Datadog positions the release as the current recommended zero-shot generation for BOOM-style infrastructure metrics.
Recommended For
- • Infrastructure, observability, and telemetry forecasting
- • Sparse, noisy, high-dimensional operational metrics
Strengths
- • Built around real observability-like workloads rather than only clean academic datasets
- • Strong benchmark fit for BOOM-style evaluation
Limitations
- • More specialized than general-purpose forecasting families
- • May be less intuitive as a default pick for simple low-dimensional business series
- • Fine-tuning and exogenous-variable support are planned upstream for Toto 2.0 but are not available in the current release
Capabilities
Tags
Specifications
- Parameters
- 4M
- Architecture
- u-muP-scaled decoder-only transformer with alternating time/variate attention and quantile output head
- Context length
- 512
- Max context
- 4,096
- Minimum history
- 32
- Recommended history
- 512
- Input step
- 32 points
- Required target series
- 1
- Temperature
- Ignored
- Top P
- Ignored
- Max output
- 2,048
- Avg latency
- n/a
- Uptime
- n/a
- Plan limits
- 1,000 rpm free · 1,000,000 rpm with billing
- Accelerator
- L40S
- Regions
- Virginia, US
- License
- Apache-2.0
Pricing
- Per forecast
- $0.00025
Performance
- Average latency
- n/a
- Availability
- n/a
- Plan limits
- 1,000 rpm free · 1,000,000 rpm with billing