No, there won't be an open attack, at least not yet. There will be more of those "hybrid" incursions. Russians will be gradually increasing the intensity of those operations, so that it becomes the new normal. There will never be a clear "well now they are attacking" boundary to point to.
Meanwhile, the EU is still buying fossil fuels from Russia, thus paying for all this. It boggles the mind.
> Given that LLMs are RL'd to the max with everything that can be given a wrong/right answer (code/maths), what currently is stopping Anthropic, OpenAI et al. from internally building (and maybe buying) datasets of good and bad architecture, and RL'ing that?
That is a question I have been asking myself. It was partially answered with Fable 5.1 and Astra. I worked with them on architectural design documents, and they were very good. Some of their suggestions were better than my solutions. They are not at a point where they can fully replace a system architect yet, but they're getting close.
There is still the open question of what to build and how to create a user interface that is understandable and presents complex information in ways understandable to humans. That's something AI cannot do yet.
This is the best font I have ever found, and I have tried a lot of them over the years. The ability to specify your own variations and generate your own fonts is really useful. I am a big fan of the dotted zero, for example.
The readability is excellent, and this font replaced everything else that I tried before. Highly recommended!
> Every time I stop reviewing the code from Claude I later find janky crap
I'm very curious about that because that has not been my experience at all. I used to find janky crap from older models and I had to review every line of code very carefully. With newer models, and especially large ones like Fable, it just doesn't happen anymore. The code that I get is what I would have written. Actually, it's usually better because it covers more edge cases that I wouldn't have thought about.
At this point, I guide Claude Code, provide general direction, and make design decisions, but I rarely write code anymore. The code that I get from the LLM is just better.
I do not fully understand where the disconnect comes from. I do believe people that say that they get crap from LLMs, and I suspect I can guess some of the reasons for that, but I cannot be sure.
> I do not fully understand where the disconnect comes from. I do believe people that say that they get crap from LLMs, and I suspect I can guess some of the reasons for that, but I cannot be sure.
Same here. There is an introspective comment below by khazhoux asking the question from the opposite experience. Why do developers end up on different sides of this coin?
LLM capability is obviously important variable. But those with bad experiences report using top tier models, so that isn't it. Likewise, level of experience can be ruled out. Each side of the coin have experienced developers reporting.
A lot of the other variables are not reported but I can go from my own experience.
Greenfield vs brownfield? The first is indeed usually easier but I get great results updating large code bases with decades of development. And I suspect developing greenfield with LLM is easier as the code is better written from the start.
Rarity of the subject matter? In principle, yet another XYZ app or web page is well represented in the training data so the LLM should have an easy go at making similar. In the past with low-end LLMs, this was noticeable. Working in relatively obscure programming languages produced just garbage. But, that was prior to the inflection point. And today I get good results working on relatively arcane scientific/mathematical algorithm as I do when I ask the LLM to barf out some web app.
Different levels of testing and validation and expectation? Even best LLMs make mistakes. The better LLMs make fewer and expose fewer still to the human as they create internal tests. Maybe people have different expectations and those with bad experience expect perfection from a single prompt? Perhaps, but any experienced developer knows bugs are a part of software development.
What else?
Whatever the answer, it's not a simple one and it may be a confluence of many smaller things. Maybe different developers have a multi-variant function in their brain that pops out a different quality-of-software score and also a different threshold that divides good from bad and all we see in their reports is that binary conclusion.
History does not support this, unfortunately. Russia has always been aggressive towards outside world and invaded its neighbors multiple times. Putin is just the latest in a long string of aggressive leaders.
People trot this out but are we saying it's an essential part of the Russian character? Couldn't we equally paint a picture of historical outward aggression of a state like Britain?
Well that's exactly my point. Instead of being fatalistic about the aggressive character of the Russian people maybe we would be better served with the mindset of inviting them into the fold of peaceful coexistence.
Here in Poland we have been inviting Russian people into the fold of peaceful coexistence for many centuries. There is a limit to how many times one can believe this to be possible.
At the beginning of war I followed a discussion on AskaRussian, with a similar approach to yours. Someone made a comment that made me think, it was along the lines of „you want our culture, but without a part about stronger dominating the weaker - that is not our culture”
Of course they didn’t represent the whole of Russia, but as anyone living next to them will tell you, the further east you go, the more „cooperation through domination” mindset there is. Win-win is a very much western thing.
But again if we went back a short time and asked people living under the domination of British imperialism then they would make similar remarks about Western culture, wouldn't they?
Are we really saying that Russians are somehow essentially warmongering in a way that other nations simply aren't? That this can't be changed?
It can be changed but they are sadly not there yet.
Historically, great defeats caused such cultures to change - look at Germany or Japan. In Russia there was a similar movement in 1991 but it sadly collapsed and they are back at it again.
Putin didn't take power in an overnight coup. Ordinary Russians didn't do anything about his growing power for what, 20 years or so? It was very much possible to act in the early stages, but it was easier not to, especially when hearing pleasing things about Russia becoming powerful.
> I know somebody who got really into blood glucose monitoring, and she tells me that the order in which she eats a meal affects how much it spikes her glucose. So if she eats a piece of toast followed by a piece of cheese, it causes her a bigger glucose spike than she'd get if she eats the cheese first and then the toast.
This is what I read in a book about glucose. So I went ahead and measured. I found it not true, at least in my case. The recommendations of eating salad before your meal do nothing measureable for my glucose spikes.
People keep praising Parakeet, but I've found it to be worse than Whisper Large. Yes, it is much, much faster and smaller, but accuracy matters a lot if you are to use dictation regularly and seriously.
I ended up having AI optimize Whisper Large and create a plugin for TypeWhisper, and that's what I use (feeding the results through local Qwen 3.8 running under MTPLX).
When you are running the model locally and trying to do other tasks at the same time, Whisper Large feels too clunky on normal hardware.
I like Parakeet because it's good enough, relatively light, and fairly fast. I'm using this for things like meeting transcription and dictation. Since I'm sending most of the text to an LLM to clean up afterwards, it works well enough.
I'm hoping for something the size of Parakeet (or smaller) but better quality. It feels like with all of the advances in making smaller models better in the LLM space, someone should be able to come up with a lightweight and better quality speech to text model.
I confirm this as my experience as well. Parakeet did not perform well for me (English with heavy accent). Whisper large is much better. But once we get to model of this size look at qwen3-asr (1.7b parameters). I'm quite happy with how it performs (both transcription quality and speed on rtx5080). By the way the model is also available on openrouter and very/very cheap. Performance is reasonable (<1s response), but some requests are delayed (>10s)
Thank you for this suggestion. I am now trying it in 8-bit quantization, and it seems to work very well, at least as well as Whisper Large. It is a little bit faster on my Mac, too.
The real verdict will be after several days of use with dual language dictation, but so far it looks really good!
Whisper large-v3-turbo is still the SOTA non-cloud model? Damn. I'm wanting to move away from it to something smaller/better but based on this comment it seems like I made the right move by sticking with whiser.cpp this whole time. It's been 2 years since large-v3-turbo was released...
It’s really domain dependent I think. If you are doing anything conversational interfacing with less AI-familiar users, latency matters a lot.
If you’re feeding the results into a very smart LLM, it will figure out what you meant (but crucially ONLY if you warn it or tell it to do so, in some cases!). If you’re writing code directly or creating something for public consumption, you can’t tolerate mistakes. If you’re taking notes for yourself you just want it to work cheaply.
If you are ok with the complexity you can run both, and a Meta/Google open model with native audio, and let a smart LLM doctor it up. If performance really matters you can pay for a proprietary model or train one yourself. Until you get to that point, I think it probably doesn't matter much either way. Voice is just too easy to fiddle with
I found Parakeet V3 TDT pretty accurate in a meeting app I build. I tried out Moonshine and whisper and found neither of them more accurate or faster than Parakeet. I also didn't like whisper take up more ram/size than parakeet.
Same thing with Nemotron 3.5 ASR. Its just as fast as parakeet and same memory footprint.
For English I've found Parakeet v2 and Parakeet unified to perform significantly better than v3. For deposition audio in a quiet room and neutral american accents I've hit 96-97% on a number of sessions with v2 and I've seen as low as 83% for heavy accents with overlapping speakers.
Meanwhile, the EU is still buying fossil fuels from Russia, thus paying for all this. It boggles the mind.
reply