Today for AI
HOT RADAR
toolsHEAT 0.0°

AI CURATED HOT STORY · 10/8/2026

Microsoft Adds Local Inference to GitHub Copilot with MAI Code 1.1 Flash MoE Model

1 reports archived0 independent sourcesupdated 10/8/2026, 06:17:08
Synthesis & Latest Updates

Microsoft announced that GitHub Copilot will support local AI model inference by the end of this month, enabling automatic or manual switching between cloud and on-device models. The update introduces MAI Code 1.1 Flash, a Mixture-of-Experts model optimized for local deployment (137B total parameters, 6B active), achieving up to 63 tokens per second decoding throughput on the Surface Laptop Ultra.

LATEST/Microsoft announced that GitHub Copilot will support local AI model inference by the end of this month, enabling automatic or manual switching between cloud and on-device models. The update introduces MAI Code 1.1 Flash, a Mixture-of-Experts model optimized for local deployment (137B total parameters, 6B active), achieving up to 63 tokens per second decoding throughput on the Surface Laptop Ultra.

HEAT TRENDHourly heat curve

163 fully observed hours
Now
0.0
Peak
11.508:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

TIMELINECoverage timeline

Total 1 reports · Latest first
  1. IT之家 · 智能时代T2·78 pts
    • GitHub Copilot adds local inference support with options for automatic orchestration or forced on-device execution.
    • Introduction of MAI Code 1.1 Flash, a specialized MoE model (137B total/6B active) using quantization and speculative decoding.
    • Achieves 40-63 tokens per second decoding throughput on Surface Laptop Ultra, significantly optimizing memory usage.