Kronos Base
onlineNeoQuasar/Kronos-base102.3M params | 512 context | $0.00025 per forecast | MIT
Kronos Base is the largest checkpoint in the NeoQuasar Kronos family and the best choice for maximum forecast quality on financial time-series workloads. It is the top of the family's quality tier, where you trade the smaller checkpoints' cheaper serving for the strongest accuracy on demanding financial forecasting.
It uses the largest tokenizer vocabulary and deepest transformer stack in the family, keeping the shared Kronos design — a GPT-style decoder over tokens produced by the financial candlestick tokenizer. Like its siblings it is designed primarily for OHLCV candlestick data; the official examples require at least OHLC columns, with volume and amount optional. The official Kronos model cards describe pretraining on over 12B K-line records from 45 global exchanges, the corpus behind the family's zero-shot coverage of market data.
On TSFM.ai reach for Kronos Base when forecast quality on financial workloads is the priority and you can absorb the heavier checkpoint. Step down to Kronos Small for a balanced middle option, or to Kronos Mini when serving cost and throughput matter more than peak accuracy.
Model Classification
Family
Kronos
Type
time series foundation model
Pretrained time-series model exposed on TSFM.ai for zero-shot or few-shot forecasting workloads.
Resources
Training Data
Official Kronos model cards describe pretraining on over 12B K-line records from 45 global exchanges.
Recommended For
- • Financial time-series forecasting with OHLCV candlestick data
- • Multi-asset price prediction across equities, forex, and crypto
Strengths
- • Purpose-built tokenizer for financial price movement patterns
- • Native OHLCV support with synthetic fallback for univariate series
Limitations
- • Domain-specific to financial data — not a general-purpose TSFM
- • Best fit when open, high, low, and close covariates are provided explicitly; the univariate fallback is a compatibility path, not the ideal contract
- • Limited public documentation and benchmark coverage compared to major TSFM families
Capabilities
Tags
Specifications
- Parameters
- 102.3M
- Architecture
- GPT-style decoder with financial candlestick tokenizer
- Context length
- 512
- Max context
- 512
- Minimum history
- n/a
- Recommended history
- n/a
- Input step
- n/a
- Required target series
- 1
- Temperature
- Supported
- Top P
- Supported
- 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
- MIT
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