6h ago

Microsoft Research AI Frontiers releases Fara1.5 computer use agent models in 4B, 9B, and 27B sizes scaling to 72% Online-Mind2Web success

27B version tops Gemini 2.5 CU and Operator on benchmarks.

0
Original post

Excited to finally share what we've been cooking! Fara1.5 is a family of computer use agents at 4B, 9B, and 27B scale. They are among the smartest, fastest, and most capable computer use models available today, competitive with even very large models. A real joy to build models that have made their way into my own workflow. Hope you find them useful too. 📝 blog: http://aka.ms/fara1.5

12:48 PM · May 21, 2026 View on X

@aravindr93 Neat!

You likely know this, but things get better with code generation actions:

Dhruv BatraDhruv Batra@DhruvBatra_

𝐈𝐧𝐭𝐫𝐨𝐝𝐮𝐜𝐢𝐧𝐠 𝐍𝐚𝐯𝐢𝐠𝐚𝐭𝐨𝐫 𝐧𝟏.𝟓 The most capable computer-use model for the web. Pareto-domination: accuracy, latency, cost • SoTA across all benchmarks • +5-10% over GPT 5.5, Opus 4.7, n1 • +25% over Gemini • 2x faster, significantly cheaper Expanded action space • UI actions (like n1) + JavaScript generation & execution

4:15 PM · May 6, 2026 · 133.8K Views
10:03 PM · May 21, 2026 · 560 Views

@DhruvBatra_ Thanks @DhruvBatra_ !

Yes ofc, your blogs are a joy to read as always 😃

I agree! Things that I think matter the most for web agents: synthetic envs; code or macro actions in general; and Dagger/on-policy distillation.

Funnily, the same things we know and love in EAI!

Dhruv BatraDhruv Batra@DhruvBatra_

@aravindr93 Neat! You likely know this, but things get better with code generation actions:

10:03 PM · May 21, 2026 · 560 Views
10:41 PM · May 21, 2026 · 61 Views

@DhruvBatra_ Also, if you have the bandwidth, would love to know the automated eval scores for n1.5

The leaderboard is taking submissions again I believe.

Dhruv BatraDhruv Batra@DhruvBatra_

@aravindr93 Neat! You likely know this, but things get better with code generation actions:

10:03 PM · May 21, 2026 · 560 Views
10:42 PM · May 21, 2026 · 42 Views