DEDUPED Hot AI repos. Real use cases. What people build. Get the Monday 5

How to run forecasts on a Mac with no GPU server with Timesfm

Use the MLX native backend to run TimesFM 3.0 locally on Apple silicon without PyTorch.

Short answer: Install the MLX extra with pip install timesfm[mlx]; then create the forecaster with TimesFM3Forecaster.from_pretrained on google/timesfm-3.0-pytorch from timesfm3.mlx; call predict with your context, a horizon and return_quantiles=True, or predict_batch for many series in one pass. All steps.

  • Data and analytics
  • Intermediate
  • 45 minutes
  • From the README
  • Claude Code
  • Codex
  • Cursor
  • Gemini CLI

platform. Contexts longer than the global context of 15,360 are truncated to their most recent points before decoding.

Who it is forAn analyst on a MacBook who does not want a cloud GPU.

The problemThe default path assumes CUDA, which a laptop does not have.

The outcomeLocal forecasts that match the PyTorch backend closely, at laptop speed.

What you need first

  • Python with pip, and `pip install timesfm[torch]`, or `pip install timesfm[mlx]` on Apple silicon
  • A CUDA GPU for the PyTorch examples, which set device to cuda
  • Your series as NumPy arrays

Steps

  1. Install the MLX extra with pip install timesfm[mlx]
  2. Create the forecaster with TimesFM3Forecaster.from_pretrained on google/timesfm-3.0-pytorch from timesfm3.mlx
  3. Call predict with your context, a horizon and return_quantiles=True, or predict_batch for many series in one pass
  4. Compare the shapes and values against the PyTorch backend numbers quoted in the README

Prompt for your agent

Using google-research/timesfm, set up the MLX backend on my Mac with `pip install timesfm[mlx]` and forecast <my series> 128 steps ahead with quantiles, reporting how long it takes.

Caveat. The published benchmark numbers are for the 330M model on an Apple M4 Max, so your speed will differ.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for google-research/timesfm and "Run forecasts on a Mac with no GPU server"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer