timesfm/timesfm-forecasting/references/system_requirements.md

System Requirements for TimesFM

Hardware Tiers

TimesFM can run on a variety of hardware configurations. This guide helps you choose the right setup and tune performance for your machine.

How Context Limits Are Determined

The max_context values in each tier are conservative recommendations based on memory-performance tradeoffs, not hard limits. TimesFM 2.5 supports up to 16,384 context points, but smaller values are recommended for most use cases.

Why 512 and 1024?

Factor	512 Context	1024 Context
Memory per 1000 series	~100 MB	~200 MB
Typical Use Case	Daily data, ~1-2 years	Daily data, ~2-3 years
Inference Speed	Faster	Moderate
Hardware	4-8 GB RAM	16 GB RAM or GPU

Memory Formula: RAM ≈ model_weights + 0.5 GB + (0.2 MB × num_series × context_length / 1000)

Where:

• model_weights = ~800 MB (TimesFM 2.5)
• context_length = your max_context value
• num_series = number of time series in your batch

You can use larger contexts if your hardware supports it:

• Up to 2048: Requires ~16 GB RAM for moderate batch sizes
• Up to 4096: Requires GPU or 32+ GB RAM
• Up to 16384: Maximum supported, requires significant memory

See Data Preparation Guide for context length recommendations by data frequency.

Tier 1: Minimal (CPU-Only, 4–8 GB RAM)

• Use case: Light exploration, single-series forecasting, prototyping
• Model: TimesFM 2.5 (200M) only
• Batch size: per_core_batch_size=4
• Context: Limit max_context=512
• Expected speed: ~2–5 seconds per 100-point series

model.compile(timesfm.ForecastConfig(
    max_context=512,
    max_horizon=128,
    per_core_batch_size=4,
    normalize_inputs=True,
    use_continuous_quantile_head=True,
    fix_quantile_crossing=True,
))

Tier 2: Standard (CPU 16 GB or GPU 4–8 GB VRAM)

• Use case: Batch forecasting (dozens of series), evaluation, production prototypes
• Model: TimesFM 2.5 (200M)
• Batch size: per_core_batch_size=32 (CPU) or 64 (GPU)
• Context: max_context=1024
• Expected speed: ~0.5–1 second per 100-point series (GPU)

model.compile(timesfm.ForecastConfig(
    max_context=1024,
    max_horizon=256,
    per_core_batch_size=64,
    normalize_inputs=True,
    use_continuous_quantile_head=True,
    fix_quantile_crossing=True,
))

Tier 3: Production (GPU 16+ GB VRAM or Apple Silicon 32+ GB)

• Use case: Large-scale batch forecasting (thousands of series), long context
• Model: TimesFM 2.5 (200M)
• Batch size: per_core_batch_size=128–256
• Context: max_context=4096 or higher
• Expected speed: ~0.1–0.3 seconds per 100-point series

model.compile(timesfm.ForecastConfig(
    max_context=4096,
    max_horizon=256,
    per_core_batch_size=128,
    normalize_inputs=True,
    use_continuous_quantile_head=True,
    fix_quantile_crossing=True,
))

Tier 4: Legacy Models (v1.0/v2.0 — 500M parameters)

• ⚠️ WARNING: TimesFM v2.0 (500M) requires ≥ 16 GB RAM (CPU) or ≥ 8 GB VRAM (GPU)
• ⚠️ WARNING: TimesFM v1.0 legacy JAX version may require ≥ 32 GB RAM
• Recommendation: Unless you specifically need a legacy checkpoint, use TimesFM 2.5

Memory Estimation

CPU Memory (RAM)

Approximate RAM usage during inference:

Component	TimesFM 2.5 (200M)	TimesFM 2.0 (500M)
Model weights	~800 MB	~2 GB
Runtime overhead	~500 MB	~1 GB
Input/output buffers	~200 MB per 1000 series	~500 MB per 1000 series
Total (small batch)	~1.5 GB	~3.5 GB
Total (large batch)	~3 GB	~6 GB

Formula: RAM ≈ model_weights + 0.5 GB + (0.2 MB × num_series × context_length / 1000)

GPU Memory (VRAM)

Component	TimesFM 2.5 (200M)
Model weights	~800 MB
KV cache + activations	~200–500 MB (scales with context)
Batch buffers	~100 MB per 100 series at context=1024
Total (batch=32)	~1.2 GB
Total (batch=128)	~1.8 GB
Total (batch=256)	~2.5 GB

Disk Space

Item	Size
TimesFM 2.5 safetensors	~800 MB
Hugging Face cache overhead	~200 MB
Total download	~1 GB

Model weights are downloaded once from Hugging Face Hub and cached in ~/.cache/huggingface/ (or $HF_HOME).

GPU Selection Guide

NVIDIA GPUs (CUDA)

GPU	VRAM	Recommended batch	Notes
RTX 3060	12 GB	64	Good entry-level
RTX 3090 / 4090	24 GB	256	Excellent for production
A100 (40 GB)	40 GB	512	Cloud/HPC
A100 (80 GB)	80 GB	1024	Cloud/HPC
T4	16 GB	128	Cloud (Colab, AWS)
V100	16–32 GB	128–256	Cloud

Apple Silicon (MPS)

Chip	Unified Memory	Recommended batch	Notes
M1	8–16 GB	16–32	Works, slower than CUDA
M1 Pro/Max	16–64 GB	32–128	Good performance
M2/M3/M4 Pro/Max	18–128 GB	64–256	Excellent

CPU Only

Works on any CPU with sufficient RAM. Expect 5–20× slower than GPU.

Python and Package Requirements

Requirement	Minimum	Recommended
Python	3.10	3.12+
numpy	1.26.4	latest
torch	2.0.0	latest
huggingface_hub	0.23.0	latest
safetensors	0.5.3	latest

Optional Dependencies

Package	Purpose	Install
jax	Flax backend	pip install jax[cuda]
flax	Flax backend	pip install flax
scikit-learn	XReg covariates	pip install scikit-learn

Operating System Compatibility

OS	Status	Notes
Linux (Ubuntu 20.04+)	✅ Fully supported	Best performance with CUDA
macOS 13+ (Ventura)	✅ Fully supported	MPS acceleration on Apple Silicon
Windows 11 + WSL2	✅ Supported	Use WSL2 for best experience
Windows (native)	⚠️ Partial	PyTorch works, some edge cases

Troubleshooting

Out of Memory (OOM)

# Reduce batch size
model.compile(timesfm.ForecastConfig(
    per_core_batch_size=4,  # Start very small
    max_context=512,        # Reduce context
    ...
))

# Process in chunks
for i in range(0, len(inputs), 50):
    chunk = inputs[i:i+50]
    p, q = model.forecast(horizon=H, inputs=chunk)

Slow Inference on CPU

# Ensure matmul precision is set
import torch
torch.set_float32_matmul_precision("high")

# Use smaller context
model.compile(timesfm.ForecastConfig(
    max_context=256,  # Shorter context = faster
    ...
))

Model Download Fails

# Set a different cache directory
export HF_HOME=/path/with/more/space

# Or download manually
huggingface-cli download google/timesfm-2.5-200m-pytorch