jascha_eng

What makes me most excited is that we probably get fable 5.5 soon with the IPO. Which is going to surpass this again. But cool to see progress!

huh? autosave, bsod, security the client side hacks of gaming how is any of this a hardware improvement. Software got a lot better.

Of course hardware also got better but software also got a let better at working around the hardwares quirks.

+1 for the omniscience benchmark!

Although even this one needs to be taken with a grain of salt. Fable/Opus still act very differently to Astra and Argon imo. But maybe it's just because I am used to these. I wouldn't clearly put them above each other though, although I still prefer anthropics models, but that could be simply out of comfort.

yeh running some automatic scripts against public urls to check if https is enabled and scanning if there are public s3 buckets is barely any better than a security linter.

There are some teams that are better than this, but they are rare.

Btw, I do work for Tiger Data/timescale, but am not at all associated with this team. I just found this release really exciting.

I sent the thread to Erik (the main contributor) though, so if you have feedback or questions, good chance that he'll see it.

Well you can always move to tiger cloud if you want it managed :) And this is PostgreSQL licensed just like our pg_textsearch which has been adopted to Googles AlloyDB and microsofts HorizonDB already, so there is hope!

benchmarks are in the blog post and the benchmarking suite is also open sourced alongside it.

from the accompanying blog post: https://www.tigerdata.com/blog/announcing-pg_vectorsearch-sc...

> TLDR: We are launching pg_vectorsearch, a new open-source PostgreSQL extension that makes vector search dramatically faster and more cost-efficient. Its new PRISM index uses partitioning and compression instead of HNSW's graph-based approach, delivering faster queries with a fraction of the memory and significantly faster index builds. At 100 million vectors, PRISM achieved 6 ms query latency on a machine with 4× less memory than HNSW, with index builds over 20× faster.

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.

-5 on omniscience? https://artificialanalysis.ai/evaluations/omniscience

That's not particularly great.

That said I love that they don't seem to restrict cyber capabilities to any degree and even lean into it.

If the best model for cyber attacks is open for everyone to use it just makes us all safer I think. Of course you then also HAVE to use it or otherwise you're vulnerable, which is a great distribution play.

Yeh it's absurd that people claim this all the time. It's some crazy conspiracy theory and when you ask for examples nothing ever shows up.

It would be economical suicide from anthropic and OpenAI to actually need models intentionally.

But hey I guess it's hard with technology that truly seems like magic. People say if you'd bring electricity to the middle ages you'd be called a witch and burned. The same is happening to the model labs here because they are bringing tech that the world isn't ready for yet.

I played around a bit with pencil.devs mcp integration before: https://www.pen.dev/

Which was at least fun for a bit although I failed to really see the use after models became better at designing themselves.

MCP is the answer but it's only really worth it if you product is reasonably complex e.g. n8n has a community MCP fully for this reason.

32 on the omniscience index. Not terrible but far from Astra and fable: https://artificialanalysis.ai/evaluations/omniscience

That's what it took to finally support Agents.md I guess. Boris had to blog about having made a mistake.

How about you stop taking things so personal instead and let others steer more.

Self hosted database access management, think Code Reviews for SQL Statements for when your devs need to go to prod: https://github.com/kviklet/kviklet

Recently cleaned up my postgres proxy and built a MySQL wire proxy so that you can use psql/datagrip as a dev but every statement is still logged. With SSO and no password sharing ofc!

But it makes it really weird to type one handed I think.

But it is actually a great model it e.g. got the carwash question right from 9 months ago. While openais models all struggled.

Fair but I think those two behaviours are strongly correlated at least the index does represent my personal experience very well where fable is way better than opus opus is better than sol. And I haven't tried Astra yet but it having 44/43 is very interesting at least.

I dont think you read my message. Muse and sol are nowhere near fable and Astra on the omniscience index

Imo the omniscience index they have has the highest correlation to actual usefulness of the models.

https://artificialanalysis.ai/evaluations/omniscience

> measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer.

This is so useful because it makes you actually trust a models output. A high score on benchmarks is not as useful because a model overtrained to always answer will give confidently wrong responses. But this index measures how often it is correct while penalizing wrong responses so that a high score means you can trust this model more and when it doesn't know it is more likely to tell you that it really doesn't know rather than making shit up.

Fable also performs a lot better than opus 5 here which correlates very strongly with perceived strength despite the models performing similarly on e.g. DeepSWE

Astra is a big jump from sol and performs the same or slightly better than fable here.

The ai hype word for 2026 after agent in 2025 for any LLM powered application.

Well kind of, I wouldn't be surprised to see that some things marketed as agents are actually good old deterministic software.

Instagram, Facebook and even threads all had much more mundane growth rates and definitely no unexpected jumps like GitHub is experiencing. I'm sure if suddenly the solar system had 10 more earths with each about 10 billion people and they would all start using Instagram tomorrow we would have exactly the same growing pains and outages that GitHub has today.

Luckily for Meta agents are not yet as much into doomscrolling as humans are.

1 in 3 is not terrible you just need a few more humans in the loop to reduce the error rate meaningfully. Combined with other classifier models and heuristics you can get good results. Humans can probably also perform better if they don't have to judge every single command but just suspicious ones our attention is limited after all.

But fables is often worth diving into at least. Opus 5 says this and then rambles on about something completely irrelevant or even wrong

The study he cites is also specifically using digital pens

> Brain electrical activity was recorded in 36 university students as they were handwriting visually presented words using a digital pen and typewriting the words on a keyboard

kinda funny.

It's also not clear from this study that there is actually any benefit to "more brain activity". Of course doing more complex motor tasks requires more brain activity but nobody guarantees that this helps with remembering things.

tbh AI is great at vague strategic decision making with a low risk bias. I wouldn't mind working for Claude