iiinova — where your digital self keeps evolving

Aymeric @aymeric.fyi · 08/09/2026

Rather than better models at similar or higher price points, I’d rather have current SOTA performance at lower prices. Imagine Opus 4.8 or 5 performance at Sonnet price level or lower. At least keep the same family of models in the same price range. Gemini 3.5 Flash Lite is over 500% since 2.0.

000

Aymeric @aymeric.fyi · 17/08/2026

I reached the same conclusion that it’s impressive to achieve this much locally but is too slow I observed one major difference in my test, it did better and faster with xhigh reasoning than medium. On my test on a M3 Max: Medium: tokens in 94k / out 78k - 2h20 XHigh: tokens in 70k / out 44k - 1h

aymeric.fyi

Qwen3.8 27B: A Free and Slow Sonnet 5?

Qwen3.8 27B just got released, a week after Meta's Muse Glimmer 30B. Alibaba advertises benchmark scores on par with Opus 4.6. This post compares Qwen3.8 27B with Muse Glimmer 30B, Sonnet 5 and Opus 5...

020

Aymeric @aymeric.fyi · 16/08/2026

I tried Qwen3.8 27B, the new local model Alibaba benchmarks against Opus 4.6. In my test, Qwen3.8 lands at Sonnet 5 quality, not the Opus 4.6 on the benchmark chart, and it needed an hour where Sonnet needed twelve minutes, on a M3 Max. Great results for a local model.

aymeric.fyi

Qwen3.8 27B: A Free and Slow Sonnet 5?

010

Aymeric @aymeric.fyi · 11/08/2026

On a M3 Max with 96GB of RAM, I compared Muse Glimmer with Qwen3.6 35B-A3B and Gemma4 26B-A4B. They achieved 20 tok/s, 70 tok/s and 60 tok/s respectively. But as Muse Glimmer is greedier, it runs through a lot more content, so it feels much 10x slower rather than only 3x.

000

Aymeric @aymeric.fyi · 08/08/2026

Here’s my Claude Code setup The post goes through install, settings, models and effort levels, and my favorite MCPs, CLIs and plugins I keep the setup fairly vanilla on purpose What I install is about letting Claude reach third-party tools: the browser, Xcode, Datadog, Linear, Google Docs, etc

aymeric.fyi

My Claude Code Setup

I have been using Claude Code personally and professionally for about six months. This post explains how I set it up, including plugins, tools, MCPs and CLI.

010

Aymeric @aymeric.fyi · 14/07/2026

With AI replacing the 1X work all around them, they are in the most powerful position they have ever been in.

000

Aymeric @aymeric.fyi · 14/07/2026

Michael Novati: Too many people are trying to become first principles thinkers […] — I’m not even sure it wants more of them. What scales is the other kind: the exceptional pattern matchers, the 10X people you’ve never heard of, happy under the radar, making things actually run.

michaelnovati.substack.com

AI Erased the Hardest Part of Being Me

I could always see the pattern. The tax was making everyone else see it too.

100

Aymeric @aymeric.fyi · 12/06/2026

I wrote my takeaways from LeadDev #LDX3 in this post. * Gains from AI are real but uneven across companies * Existing bottlenecks in the SDLC become blockers in AI-enabled orgs * Human relationships are still very important to build large successful products

aymeric.fyi

LeadDev LDX3 London 2026 Summary

What 25 talks and panels at LDX3 London 2026 said about engineering teams in the AI era.

000

Aymeric @aymeric.fyi · 04/06/2026

Interesting, thanks for sharing. Based on the README, I’m wondering if we could achieve the same set of features with Ansible roles & playbooks.

100