The Circuitry
THE CIRCUITRYYour one-stop source for all tech news
  • Today
  • Feed
  • Events
  • Saved
HOMETODAYFEEDEVENTS
BOOKMARKS
RSS
© 2026 The Circuitry
About UsSourcesContactCorrectionsPrivacy
  • Today
  • Feed
  • Events
  • Saved
  • AAPL
  • MSFT
  • GOOGL
  • META
  • NVDA
  • TSLA
  • AMZN
  • INTC
  • SONY
  • TTWO
  • COIN
  • BTC
  • ETH
  • SOL
  • AAPL
  • MSFT
  • GOOGL
  • META
  • NVDA
  • TSLA
  • AMZN
  • INTC
  • SONY
  • TTWO
  • COIN
  • BTC
  • ETH
  • SOL

Feed

Ad
From The CircuitryVerified tech news, cross-checked.How we verify →
  • OpenAI cuts reasoning-extraction success to 0.2% on GPT-6.2 AstraTech
  • OpenAI introduces MedChemBench, with GPT-6.2 Astra at 84.5%Tech
  • Cairn agents found a zero-day and took control of a serverTech
  • TikTok launches an AI checker and starts a global likeness rolloutTech
  • NVIDIA's third-quarter revenue rises 103% to $115.8 billionTech
  • Falcon 9 launches NROL-136 and lands the boosterTech
  • SpaceX confirms Dragon splashdown, completing its 35th resupplyTech
  • Falcon 9 launches USSF-317 and lands the boosterTech
  • Anthropic mathematician says he believes he has a Riemann proofTech
  • Google will retire the Nest app and Assistant on home devices in 2027Tech
  • Niron Magnetics will build a $1.8 billion plant in DothanTech
Scroll for more
Verification
VERIFIEDConfidence: HIGH
Source identified
Claims cross-referenced
No discrepancies found
Fact-check summary

Microsoft's official October 7 announcements confirm local MAI-Code-1.1-Flash for Windows PCs and GitHub Copilot on high-end hardware. Corrected Oct. 7: memory figure now matches Microsoft's Command Line post 'Bringing local models and sandboxed tools to Windows and GitHub Copilot' (peak 75.5 GB at 256K context, 53 GB quantized); the earlier 'more than 120 GB' recommendation came from Frandroid, not Microsoft. Corrected Oct. 7 (2): Claude Opus 5.5 (87.64%) and GPT-6 Astra (87.27%) Terminal-Bench 2.1 scores are from Vals AI's leaderboard (vals.ai/benchmarks/terminal-bench-2-1, updated 9/28/2026, Terminus 2 harness), not Microsoft; now attributed.

Sourcing
1source
Frandroid · track record
63Stories
100%Verified
130d
All sources →
Markets
MSFT···

Live quote · not investment advice

Ad
From The CircuitryWhy The Circuitry

Verified tech news, cross-checked.

Every story is checked against independent sources before it posts — no rumors dressed up as fact.

How we verify →
From The CircuitryWhy The Circuitry

Verified tech news, cross-checked.

Every story is checked against independent sources before it posts — no rumors dressed up as fact.

How we verify →
Home/Tech/Microsoft preparing local MAI-Code-1.1-Flash for high-end PCs
VERIFIEDBy Xavier Rivera·Wednesday, October 7, 2026 at 4:19 PM AST·1.5 min read

Microsoft preparing local MAI-Code-1.1-Flash for high-end PCs

Microsoft has introduced a local edition of MAI-Code-1.1-Flash that runs on Windows PCs without server access. The model requires substantial memory and is coming to GitHub Copilot in experimental preview by the end of October, starting with Nvidia RTX Spark PCs.

Source:Frandroid
Post
Microsoft preparing local MAI-Code-1.1-Flash for high-end PCs
Ad
From The CircuitryVerified tech news, cross-checked.How we verify →
Ad
From The CircuitryWhy The Circuitry

Verified tech news, cross-checked.

Every story is checked against independent sources before it posts — no rumors dressed up as fact.

How we verify →
TL;DRAI · 60 sec read

Microsoft introduces a local MAI-Code-1.1-Flash coding model that runs on Windows PCs without servers. It comes to GitHub Copilot in experimental preview by the end of October, starting with Nvidia RTX Spark PCs, and incurs no token costs. Memory use peaks at 75.5 GB at full 256K context, so it effectively needs a 128 GB RTX Spark machine. It trails leading cloud models such as Claude Opus 5.5 on Vals AI's Terminal-Bench 2.1 leaderboard.

Morning Brief

Get the daily Microsoft brief

Every morning, in your inbox. What's in The Brief →

⚐ CORRECTEDWednesday, October 7, 2026 at 6:57 PM EDT

Wrong: Claude Opus 5.5's 87.6% and GPT-6 Astra's 87.3% Terminal-Bench 2.1 scores read as Microsoft's own figures. Right: those scores come from Vals AI's Terminal-Bench 2.1 leaderboard, which uses a different test setup; Microsoft reported only its own model's scores (62.9% cloud, 66.3% local).

EARLIER CORRECTIONS (2)
Wednesday, October 7, 2026 at 6:34 PM EDT

Wrong: said Microsoft recommends more than 120 GB of video memory to run the local MAI-Code-1.1-Flash. That figure came from Frandroid, not Microsoft. Right: Microsoft says the local model's memory use peaks at 75.5 GB at its full 256K context, so it effectively needs a 128 GB RTX Spark machine.

Wednesday, October 7, 2026 at 6:19 PM EDT

Wrong: said MAI-Code-1.1-Flash activates 5 billion parameters per step and was "confirmed in development" for GitHub Copilot, and gave only the cloud model's 62.9% Terminal-Bench 2.1 score. Right: Microsoft says the model has 137 billion total and 6.8 billion active parameters; the local version comes to GitHub Copilot in experimental preview by the end of October; on Terminal-Bench 2.1 the cloud model scored 62.9% and the local version 66.3%.

→ All corrections
Microsoft has introduced a local version of its MAI-Code-1.1-Flash AI coding model that runs directly on Windows PCs without using company servers.

The local model comes to GitHub Copilot by the end of October. Microsoft says the local edition arrives in experimental preview in GitHub Copilot, starting with PCs equipped with Nvidia RTX Spark chips. Computations occur on the device itself, so Microsoft states users will not be billed for tokens.
Microsoft says memory use peaks at 75.5 GB at the full 256K context, so it effectively needs a 128 GB RTX Spark machine.

Hardware demands restrict availability to high-end machines. The quantized local model takes up 53 GB, and Microsoft says its memory use peaks at 75.5 GB at the full 256K context, so it effectively needs a 128 GB RTX Spark machine, such as the Surface Laptop Ultra with 128 GB of unified memory.

Performance trails leading cloud models on Terminal-Bench. Microsoft reports a 62.9 percent success rate on Terminal-Bench 2.1 for the cloud model and 66.3 percent for the local version. On Vals AI's Terminal-Bench 2.1 leaderboard, which uses a different test setup, Claude Opus 5.5 scores 87.6 percent and GPT-6 Astra scores 87.3 percent. On SWE-Bench Verified the local version reaches 70.8 percent while the cloud version reaches 72.6 percent.
From The CircuitryThe Feed — live briefs across tech, all day.See what’s happening →

Model uses mixture-of-experts architecture. MAI-Code-1.1-Flash contains 137 billion total parameters yet activates only 6.8 billion at each step. The model first appeared on August 11 and entered GitHub Copilot this summer at $0.20 per million input tokens and $1.20 per million output tokens.
Computations occur on the device itself, so Microsoft states users will not be billed for tokens.

Local execution eliminates token costs. Microsoft positions the offline capability as an alternative to cloud services such as Claude Haiku 5.5 and GPT-6 Luna. The company continues broader work to enable large models to run locally on Windows.
Ad
From The CircuitryWhy The Circuitry

Verified tech news, cross-checked.

Every story is checked against independent sources before it posts — no rumors dressed up as fact.

How we verify →
Why this mattersAI · ~100 words

Tap a lens to see what this story means for you.

Morning Brief

Liked this? The Brief brings you the whole day in tech, verified, every morning.

Two minutes, free forever. What's in The Brief →

Reader-supported
DonateBuy me a coffee →Follow@thecircuitry_ →Follow@thecircuitry.to →
HELP US IMPROVE
Grand Theft Auto VILive hubUS store unlocks Thu Nov 19, 1:00 AM ASTFirst unlock: New Zealand · Wed 7:00 AM ASTLaunch hub →Image: Rockstar Games
From The Circuitry

See what’s happening right now

The Feed runs all day — short, verified briefs the moment they break.

Open the Feed →
From The Circuitry

Follow @thecircuitry_

Every story we publish, as it happens. No noise between.

Follow on X ↗On Bluesky ↗

Reader-supported

The Circuitry is a passion project I've always wanted to build, and I love the work behind it.

Running it costs real money. APIs, hosting, time. To keep improving the site and growing this into something useful for everyone, those costs have to be covered.

Any contribution is appreciated. If not, no pressure. Thanks for reading.

Buy me a coffee
microsoftaigithub-copilot
Grand Theft Auto VILive hubUS store unlocks Thu Nov 19, 1:00 AM ASTFirst unlock: New Zealand · Wed 7:00 AM ASTLaunch hub →Image: Rockstar Games
More inTech
  • Falcon 9 launches NROL-136 and lands the booster

    Tech · 1h
  • SpaceX confirms Dragon splashdown, completing its 35th resupply

    Tech · 16h
  • Falcon 9 launches USSF-317 and lands the booster

    Tech · 1d
SupportThe Work

The Circuitry is reader-supported. If you find the daily brief useful, you can buy me a coffee to keep it going.

Buy a coffee →
Ad
From The CircuitryWhy The Circuitry

Verified tech news, cross-checked.

Every story is checked against independent sources before it posts — no rumors dressed up as fact.

How we verify →
From The CircuitryWhy The Circuitry

Verified tech news, cross-checked.

Every story is checked against independent sources before it posts — no rumors dressed up as fact.

How we verify →

MORE IN THIS BEAT

All Tech →
  • Tech·Nov 20

    OpenAI cuts reasoning-extraction success to 0.2% on GPT-6.2 Astra

    OpenAI has cut the average success rate of attacks that extract protected reasoning from an earlier request to 0.2% on GPT-6.2 Astra, from 54.0% on GPT-6 Astra. Current-request attacks are at 0.9%, and selected safeguard improvements are being applied back to GPT-6 Astra.

  • Tech·Nov 20

    OpenAI introduces MedChemBench, with GPT-6.2 Astra at 84.5%

    OpenAI has introduced MedChemBench, a 99-problem test of how models read molecules, explain lab results, and choose the next compound. At the xhigh setting, GPT-6.2 Astra scored 84.5%, while compound choice and property forecasts remain much weaker than structure work.

  • Tech·Nov 20

    Cairn agents found a zero-day and took control of a server

    Autonomous Cairn agents found an unpatched flaw in a medical imaging app and took control of a server in a university campaign that began September 8, 2026. Amazon rates a Chinese-nexus link as moderate confidence, and the project advisory still lists no patched version.

  • Tech·Nov 20

    TikTok launches an AI checker and starts a global likeness rollout

    TikTok has launched a public AI checker for images and videos made with its own tools and has started a global opt-in likeness rollout. The checker is live without an account, while the likeness tool is limited to eligible creators and a notice does not by itself remove a post.

  • Tech·Nov 17

    NVIDIA's third-quarter revenue rises 103% to $115.8 billion

    NVIDIA has reported third-quarter revenue of $115.8 billion, up 103% from a year ago, with data-center revenue of $107.6 billion. Fourth-quarter revenue is expected at $130.0 billion, plus or minus 3%, with no data-center compute revenue from China assumed.