No wonder they don't publish an availability percentage. If I was a business customer paying for GitHub I would be very upset with the availability lately.
Someone built an archive of Github statuses to show aggregate uptime, last month and this month Github's uptime is below 90%, not even one "nine" of availability: https://mrshu.github.io/github-statuses/
87% uptime for Github in February 2026. They've got to get it together.
Did gemini-2.5-flash-image get an upgrade as well? I just got the following, which is fascinating, and not something I've seen before:
> I'm sorry, but I cannot fulfill your request as it contains conflicting instructions. You asked me to include the self-carved markings on the character's right wrist and to show him clutching his electromancy focus, but you also explicitly stated, "Do NOT include any props, weapons, or objects in the character's hands - hands should be empty." This contradiction prevents me from generating the image as requested.
My prompts are automated (e.g. I'm not writing them) and definitely have contained conflicting instructions in the past.
A quick google search on that error doesn't reveal anything either
Most production software is wrappers around existing libraries. The relevant question is whether this wrapper adds operational or usability value, not whether it reimplements OCR. If there are architectural or reliability concerns, it’d be more useful to call those out directly.
That’s weird, pnpm no longer automatically runs lifecycle scripts like preinstall [1], so unless they were running a very old version of pnpm, shouldn’t they have been protected from Shai-Hulud?
Let me understand it fully. That means they updated dependencies using old, out of date package manager. If pnpm was up to date, this would no have happened? Sounds totally like their fault then
Does anyone here understand "interleaved scratchpads" mentioned at the very bottom of the footnotes:
> All evals were run with a 64K thinking budget, interleaved scratchpads, 200K context window, default effort (high), and default sampling settings (temperature, top_p).
I understand scratchpads (e.g. [0] Show Your Work: Scratchpads for Intermediate Computation with Language Models) but not sure about the "interleaved" part, a quick Kagi search did not lead to anything relevant other than Claude itself :)
AFAICT, kimi k2 was the first to apply this technique [1]. I wonder if Anthropic came up with it independently or if they trained a model in 5 months after seeing kimi’s performance.
Anthropic is encouraging the "have the model write a script" technique as well, buried in their latest announcement on Claude Agent SDK, this stuck with me:
> The Claude Agent SDK excels at code generation—and for good reason. Code is precise, composable, and infinitely reusable, making it an ideal output for agents that need to perform complex operations reliably.
> When building agents, consider: which tasks would benefit from being expressed as code? Often, the answer unlocks significant capabilities.
I'm doing coreference resolution and this model (w/o thinking) performs at the Gemini 2.5-Pro level (w/ thinking_budget set to -1) at a fraction of the cost.
Awesome, I've been playing in the ebook space myself, will check it out. Particularly interested in digging into the code too see how you skip headers, footnotes, etc.
Just one quick note as I ran into this when setting it up:
╰─▶ Because the requested Python version (>=3.8) does not satisfy Python>=3.10,<3.13 and kokoro==0.9.4 depends on Python>=3.10,<3.13, we can conclude that kokoro==0.9.4 cannot be used.
Note I definitely disregarded your instructions and used `uv` to setup the project. Still, it seems like changing the `pyproject.toml` to `requires-python = ">=3.10"` would be good considering kokoro's Python version support.
reply