htrp

6 ms·

91.
by htrp 2mo ago
> Hugging Face published a timeline that said the rogue agent broke into a sandbox — a testing environment — “hosted on a third-party provider’s infrastructure” and then launched its broader hack from there, Reuters said. Hugging Face di
92.
by htrp 2mo ago
how bad does the app burn battery life for your watch?
93.
by htrp 2mo ago
math notation doesn't bias towards English language understanding like pseudocode
94.
by htrp 2mo ago
Token Type Price per 1M tokens Cached Input $0.27 Input $2.70 Output $13.50 Looks like they are first vendor to undercut in price.
95.
by htrp 2mo ago
Fired their pre-training team, shutting down Nova, shutting down the Adept Acqui-hire but still building another Foundation model >According to the people familiar with the matter, resources have increasingly moved away from the existing
96.
by htrp 2mo ago
Why is everyone's pricing the same? licensing agreements with moonshot as an inference provider?
97.
by htrp 2mo ago
> which OpenAI seems to allow via API but not Anthropic Does openai still allow logprobs in their current gen models?
98.
by htrp 3mo ago
> And no, no AI company has ever come to us and asked to run training on all of our scanned copies Yet
99.
by htrp 3mo ago
We actually tracked this over the last year, bedrock is significantly better than the anthropic direct endpoints
100.
by htrp 3mo ago
xAI did lose all of their founders and they are rebooting the company because it was "built wrong"
101.
by htrp 3mo ago
OpenAi cries in sora
102.
Stripe in talks to buy OpenRouter for ~10B(wsj.com)
62 points by htrp 3mo ago|25 comments
103.
by htrp 3mo ago
I always thought it was a pretty noisy signal, was there any empirical work showing what the SNR was in these step by step classification loops?
104.
by htrp 3mo ago
still removing work from github. this is formalizing some very enterprise-esque processes for security research, software resellers anyone?
105.
by htrp 3mo ago
Any updates here to make this work on more modern llms?
106.
by htrp 3mo ago
>The number traces to Meta’s Llama 3 technical report, which documented 419 unforeseen disruptions across 16,384 H100s over 54 days of training, of which 148 were GPU failures and 72 were HBM3 memory failures. From an annualized number o
107.
by htrp 3mo ago
I'm seeing it for 4k and up for similar specs as the spark?
108.
by htrp 3mo ago
>The browser swarm from earlier this year peaked at roughly 1,000 commits per hour on Git. The new system peaks at around 1,000 commits per second. >To facilitate this rate of activity, we built a new version control system (VCS) from
109.
by htrp 3mo ago
> This is a very strange article considering that Llama, the mother of all open-weight models, has led to anything but success for Meta. The llama drama will be a netflix show of it's own in 5 years.
110.
by htrp 3mo ago
Don't use Opencode, Don't use remote models (cloud providers), Don't use Docker to isolate coding agents. May as well write a post saying don't use LLM's for any SWE work. >Conclusion Stop using OpenCode. >Pos
111.
US Considers Creating Finra-Like Watchdog to Vet Top AI Models(bloomberg.com)
4 points by htrp 3mo ago|0 comments
112.
by htrp 3mo ago
another usage reset incoming i guess
113.
by htrp 3mo ago
deepmind using AI to evaluate submissions?
114.
by htrp 3mo ago
>Alphabet Inc.’s Google is months behind schedule on delivering Gemini 3.5 Pro, its most powerful flagship AI model, because the company has been taking time to try to improve its capabilities, particularly in coding, according to people
115.
by htrp 3mo ago
>The story of Reflection AI is supposedly that the company was faffing and failing at winning in the coding agent space, but was introduced to Jenson, who suggested they build an open-weight model and said he would fund it. That turned i
116.
Inkling – Open-Weights 975B Parameter LLM(thinkingmachines.ai)
121 points by htrp 3mo ago|4 comments
117.
by htrp 3mo ago
https://thinkingmachines.ai/model-card/inkling/ 975B parameter 41B active
118.
by htrp 3mo ago
This just means that DC builds will move to other states. It isn't exactly like you need low latency/colocation for AI workloads.
119.
by htrp 3mo ago
Is this because your free customers bring their own ai/tokens?
120.
by htrp 3mo ago
how does this compare with the aws lambda microvms?

More ›