b9932
vulkan: disable FA mask_opt on GCN to improve performance (#24362) vulkan: disable FA mask_opt on GCN to improve performance reenable mask opt over attention head size…
Full feed
Feed 28,116 posts
Every story across every category, newest first. Each card links to the original publisher; daily-brief posts open as editorial pages.
RT Jarred Sumner: Rewriting Bun in Rust https://bun.com/blog/bun-in-rust
RT Jarred SumnerRewriting Bun in Rusthttps://bun.com/blog/bun-in-rust
RT Déborah: Wow amazing Grok Imagine 😍
RT DéborahWow amazing Grok Imagine 😍
Open Source Barware: free, local-first bar inventory software (GPLv3)
Article URL: https://opensourcebarware.com Comments URL: https://news.ycombinator.com/item?id=48837809 Points: 4 # Comments: 0
Suspecting AI cheating, Ivy League prof ordered an in-person final; scores fell 50%
AI cheating leads to "a failed society," professor says.
Grok has always been very strong on law
Grok has always been very strong on lawTech Dev Notes: Grok 4.5 is #1 on Harvey's Legal Agent Benchmark
Rate of improvement is 🚀🚀
Rate of improvement is 🚀🚀Aman Sanger: Very excited for this model. It’s an enormous improvement over composer 2.5 and trained entirely from scratch. It’s been a…
RT sourcery: .@Starcloud_ CEO @PhilipJohnston: "Buy SpaceX stock." "It's like owning the railroads." " The SpaceX IPO is going to be viewed historical…
RT sourcery.@Starcloud_ CEO @PhilipJohnston: "Buy SpaceX stock.""It's like owning the railroads."" The SpaceX IPO is going to be viewed historically as the most undervalued IPO of…
Rate of improvement is accelerating. Users should notice a meaningful improvement in the usefulness of the Grok Build harness with our V9 foundation m…
Rate of improvement is accelerating. Users should notice a meaningful improvement in the usefulness of the Grok Build harness with our V9 foundation model (aka Grok…
> Like it not, @Zai_org has set the standard for Chinese AI models. Beat it or you are not relevant, bc the next GLM model will be a lot better. No. I…
> Like it not, @Zai_org has set the standard for Chinese AI models. Beat it or you are not relevant, bc the next GLM model will…
Grok 4.5 is not yet using our internally developed C/C++ inference software that exact maps to the GB300 hardware. Doubling or more of the current spe…
Grok 4.5 is not yet using our internally developed C/C++ inference software that exact maps to the GB300 hardware. Doubling or more of the current speed…
> SWE-bench pro is saturated at 70% we knew bros
> SWE-bench pro is saturated at 70%we knew bros
Optimum number of items to inspect before buying one
Simple mathematical models are almost always too simple to model complex phenomena in the real world, and the one I want to discuss in this post…
Free will as a model parameter
The most popular take on the standard free will debate is that you are the algorithm. Your preferences and reasoning that determine your actions IS free…
the funny thing is that the old Grok/xAI team was full of high-profile people with mythical CVs, and the new one is, no offense, some young bros (rela…
the funny thing is that the old Grok/xAI team was full of high-profile people with mythical CVs, and the new one is, no offense, some young…
RT SMX 🇺🇸: How it feels to use Grok 4.5 😎
RT SMX 🇺🇸How it feels to use Grok 4.5 😎
b9931
opencl: ragged-tile MoE prefill FP16 GEMM optimization (skip padded expert tiles) (#25433) opencl: ragged-tile MoE prefill GEMM (skip padded expert tiles) The MoE prefill GEMM groups…
Introducing: Granular job tracking and cost attribution in LlamaParse 🦙 Parsing thousands of pages a day is a lot to manage. Engineering teams need…
Introducing: Granular job tracking and cost attribution in LlamaParse 🦙 Parsing thousands of pages a day is a lot to manage. Engineering teams need to have…
Find funding, fast
Some AI safety funders can take months to decide; others confirm in days. I’ve been on both sides of the grant application and know how crucial…
Grok 4.5 true usefulness is excellent
Grok 4.5 true usefulness is excellentSnorkel AI: We obtained early access to evaluate @SpaceXAI's newest Grok 4.5 model on GDPval+: real professional deliverables (from a variety…
Grok 4.5 context window will upgrade to 1M probably by next week
Grok 4.5 context window will upgrade to 1M probably by next weekArtificial Analysis: SpaceXAI’s Grok 4.5 scores 54 to place fourth on the Artificial Analysis Intelligence…
OpenAI no longer recommends SWE-Bench Pro
Article URL: https://openai.com/index/separating-signal-from-noise-coding-evaluations/ Comments URL: https://news.ycombinator.com/item?id=48837396 Points: 10 # Comments: 1
➤ Pricing of $2/$6 per 1M tokens of input/output; cache hits are discounted by 75% to $0.5 per 1M tokens, and costs still double with long (>200k tok…
➤ Pricing of $2/$6 per 1M tokens of input/output; cache hits are discounted by 75% to $0.5 per 1M tokens, and costs still double with long…
ha ha… it's sad that it is funny
ha ha…it's sad that it is funny永雏塔菲: @teortaxesTex a ds employee privately joked they're training a 15t model. i'm guessing that's where leo got it