601–610 of 616 posts
Earlier quoted context omitted.
Knowledge Graph + automated transcriptions of almost every YouTube video = giant untapped moat of data
It's not exactly untapped, my AI company has scraped YouTube transcripts for 3 years now for RAG
Earlier quoted context omitted.
And I know many people that don't. That have a 20 or 100 dollar subscription for very bursty workflows with months where they barely use tokens. Not every subscriber is a full time SWE. In fact most professional SWEs will be on enterprise plans and thus not get subscriptions at all. I think it's very plausible that subscriptions are overall losing money. But we simply don't know.
Claude Code has enterprise subscriptions. No $200 max plan but you can buy $100 premium seats. (and if you are using per-token billing as an enterprise user without first maxing out a premium seat you are very silly)
Earlier quoted context omitted.
A big slice of a small pie does not prove the majority. And I'm speaking as someone who is indeed a user of those OpenRouter Chinese models.
this is the first mention of a 'majority' so certainly looks like a moving goalpost
Pelican svg and a near-perfect 3D MacBook at max effort for $0.16, about a fifth of Fable's price. Fable 5 still wins on detail with no visible errors, but it's close. And this isn't a memorized pelican; https://playcode.io/blog/macbook-svg-benchmark#gemini-3-6-fl...
I don't know if it's a rendering error since it looks like your site renders the SVG instead of hosting a static image of it, but the 3.6 macbook looks like an abstract art piece lol, both ff and chrome desktop
Earlier quoted context omitted.
One only hopes "a few people" remains a comforting description. Consensus has an unfortunate habit of beginning that way.
Have you created other burner accounts to reply to me in the past, or is this your first one?
And outrage at the unimpressed is.. a curious standard for a sponsored blogger.
Pelicans for 3.6 Flash and 3.5 Flash-Lite (Cyber isn't available to me through the API yet.) https://tools.simonwillison.net/markdown-svg-renderer#url=ht...
Does this say anything about the model? I meant the underlying attention/pattern it took for Gemini 3.6 Flash to create this SVG.
Earlier quoted context omitted.
I don't think so. According to some very basic research there are around 8bn searches a day, or 250bn a month. Let's assume Google serves AI overviews on every SERP (they don't) and don't cache them (they do, afiak). And let's assume that each AI overview is 2000 tokens (blended input/output), that's 500T tokens a month. It's rumoured that anthropic is serving somewhere close to 10Q tokens a month. Now it may be that…
Well I asked the google AI mode thing what it thinks about your comment and it told me this (edited obviously): "10 Quadrillion tokens a month means: 333 Trillion tokens per day and 3.85 Billion tokens generated/processed every single second, 24/7." "At an incredibly cheap, subsidized infrastructure cost of $1 per million tokens, serving 10 Quadrillion tokens would cost Anthropic $10 Billion per month ($120 Billion a…
They definitely cache results - I've searched and re-searched an identical query back to back a few times and seen identical results from overview. They are definitely throwing a stupid amount of compute towards these ai results nobody is paying for - changing punctuation and stuff does get you a different response - but they're not doing no caching.
Certainly what they're doing with their infrastructure is impressive but it's not super meaningful at the end of the day for a for profit company to be really impressively good at burning tens of billion dollars on a service nobody pays for while the same tech from their competitors is quickly becoming one of the largest spend categories for many software engineering teams
Earlier quoted context omitted.
It's also very possible that they know their big model underperforms chatgpt 5.6 and fable by too much, so they are focusing on what they can get wins in like speed instead.
I personally doubt that. It would be a shame if they cannot beat Kimi K3 or Qwen3.8 Max, both of which are claimed to be Fable-like. If that is true, it will be [or would be] the first time a major American lab falls behind a Chinese competitor.
I wonder how big the Pro model is that Google is using behind the scenes to train these smaller ones. Going on baseless speculation, the lack of accompanying pro models with these flash releases either means: 1) the model is too big to be economical, 2) google doesn't have the compute to serve the big model, 3) their big model has too many alignment issues to serve to the public. edit: looks like benchmarks are up on…
I wonder if the broad use of AI overviews on Google search results is having an impact. Maybe the numbers make it more profitable to use their compute on several billion searches a day rather than selling API access.
Earlier quoted context omitted.
Because AA Coding "Index" consists only of two benchmarks (Terminal-Bench v2.1, SciCode) and generally fails to be meaningfully representative of agentic coding capabilities.
AA coding index has been updated to use DeepSWE, Terminal-Bench v2, and SWE-Atlas-QnA.