> using another unaligned model to rewrite reasoning and then feeding it back into the model

Indeed, would be interested to understand how you do that. And where do you run the model? On a local or a rented system and what are the specs?

glm 5.2 is very unaligned, you can just use any provider with a trivial jailbreak.

Selfhosted or one of the listed? If external/hosted, which one did you use?

Congrats with your product, hope it works out well. Nice competitor analysis & (future) features list at https://github.com/KrushnaVardhanReddy/PaperPilot/blob/main/..., IMHO easier to read than your above list.

Question: seemingly you’re using Antigravity, an harness not many use. Please tell use about the choice of LLM and your experience with it.

Thanks for checking out the docs! We used a hybrid workflow: Antigravity IDE paired with Jules for asynchronous sub-tasks (like spinning up isolated unit tests and docs in parallel containers).

Having a strict test-first harness where every PR has to run against real PDF byte fixtures (zero mocks) and compile across CLI, WASM, and MCP kept the codebase consistent even when moving fast.

Thanks for your reply. Will look into Jules [0] [1].

Are you using Antigravity IDE v2.5.5 that hasn't been updated since 2026-08-13 and that supports 3 legacy models + Gemini Flash 3.8? I assume you're using Gemini Flash 3.8?

[0] https://news.ycombinator.com/item?id=44034918 [1] https://news.ycombinator.com/item?id=44813854

Netflix works fine with the device directly connected to the TV. With a long hdmi-cable (5m+) + an usbc-to-hdmi converter you can keep de phone/pad in your hands. Using a short (≤2m) hdmi-usbc cable you probably need to put the device next to the TV. I bring either when traveling to watch TV in hotels rooms.

FWIW, a Nintendo Switch USB-C to HDMI cable with power input travels well for this, albeit a bit bulky.

This one puts the charger input at the HDMI end where the TV and presumably outlets are, so power comes to your video playing device over the USB-C at your device end.

https://www.walmart.com/ip/USB-C-HDMI-Cable-Adapter-Nintendo...

In 1955, Zenith introduced the Flash-matic: world's first wireless remote control, so people wouldnt need to have a really long wire connecting their remote to their television.

I guess that could "work" but it sounds like a bit of a hassle. Perhaps an old iPad could be used.

CloudNativePG supports Kubernetes Local Storage as one of its main features [0]

  Persistent volume management, including:
    Support for Local Persistent Volumes with PVC templates.
    Reuse of Persistent Volumes storage in Pods.
    Separate volumes for WAL files and tablespaces.

Chapter Temporary PostgreSQL Cluster Degradation [1] describes how to set it up and starts with this

> While the standard approach ensures service reliability and leverages Kubernetes' self-healing capabilities, there are scenarios where operating with a temporarily degraded cluster may be acceptable. This is particularly relevant for PostgreSQL clusters relying on node-local storage, where the storage is local to the Kubernetes worker node running the PostgreSQL database. Node-local storage, or simply local storage, is employed to enhance performance.

[0] https://cloudnative-pg.io/docs/devel/#main-features [1] https://cloudnative-pg.io/docs/devel/kubernetes_upgrade/#tem...

https://artificialanalysis.ai/models/mistral-large-4 for the main stats

                          Inte         Cost
                          llig          per
  Open Weight model       ence  Speed  Task

  Mimo-V.26-Pro            46     47   $0.13
  GLM-5.3 (max)            45     73   $2.01
  DeepSeek 4.1 Flash Max   39    227   $0.27
  Mistral Large 4 Preview  38    116   $1.13

Imo omniscience correlates better to how useful the model is in practice than the intelligence index. But you have to use both together of course.

Too late to edit, now including AA-omni Score [0] and 'by Domain' Software Engineering [1]. Also added comparison to the eyeballs-median of the 'top 10' models and then you see that indeed Mistral Large 4 Preview scores miserable in the AA-Omni indices.

                          Inte         Cost     AA-   Omni
                          llig          per    Omni  Softw
  Open Weight model       ence  Speed  Task   score    Eng

  Mimo-V.26-Pro            46     47   $0.13      8     33
  GLM-5.3 (max)            45     73   $2.01     14     37
  DeepSeek 4.1 Flash Max   39    227   $0.27     -5     34
  Mistral Large 4 Preview  38    116   $1.13     -5      5

  Closed/proprietary      ~50   110-  $1.50-    ~43    ~85
     median top 10               242   $7.50

[0] https://artificialanalysis.ai/evaluations/omniscience [1] https://artificialanalysis.ai/evaluations/omniscience?detail...

Then this is something for you https://shop.digitalcourage.de/themen/ak-vorrat/lichtbildaus.... Either get it online, or buy at events like Chaos Computer Congress.

And normal people who can not apply for access to daybreak, should we use Abliterated GLM-5.3 ? [0] [1] [2]

[0] https://news.ycombinator.com/item?id=49605691 [1] https://news.ycombinator.com/item?id=49897075 [2] https://huggingface.co/models?other=offensive-security