up6w6

6 ms·

1.
Haptic Touchpad Support Expected for Linux 6.18(phoronix.com)
2 points by up6w6 1y ago|0 comments
2.
by up6w6 1y ago
crazy how they only show benchmark results against their own models
3.
Adobe launches a new 'computational photography' camera app for iPhones(theverge.com)
2 points by up6w6 1y ago|0 comments
4.
Project Indigo – a computational photography camera app(research.adobe.com)
3 points by up6w6 1y ago|1 comments
5.
by up6w6 2y ago
I am very suspicious of the results. A few months ago they published a LLM benchmark, calling it "perfect" while it actually contained like only 50 inputs (academic benchmark datasets usually contain tens of thousands of inputs).
6.
X evades Brazil ban using Cloudflare(bbc.com)
6 points by up6w6 2y ago|2 comments
7.
by up6w6 2y ago
Very audacious to call it "almost perfect" when it has only what appears to be 50 questions. For comparison, MMLU contains 57 tasks and more than 100k questions.
8.
by up6w6 2y ago
I remember that you can detect the "curl | bash" server side and serve a different script than what the user would get by downloading using other methods[1]. But yeah, the binary itself already has enough attack surface. [1] http
9.
by up6w6 2y ago
Not only that, but Servo is only receiving less 3000 USD/month in donation, less than their goal of 10k/month, and much less than what they deserve if consider Ladybird is receiving millions. https://servo.org/blog
10.
Android's new antitheft feature is now in testing(theverge.com)
1 points by up6w6 2y ago|0 comments
11.
by up6w6 2y ago
Does anyone knows what is the current status of Apple silicon hardware emulation in qemu?
12.
by up6w6 2y ago
Related news: Servo Web Engine Continues Advancing But Seeing Just $1.6k In Monthly Donations https://www.phoronix.com/news/Servo-Engine-May-2024
13.
Servo Web Engine Continues Advancing but Seeing Just $1.6k in Monthly Donations(phoronix.com)
4 points by up6w6 2y ago|0 comments
14.
by up6w6 2y ago
> Lack of speed and ghosting felt like it made traditional Eink impossible to do most computing tasks. So we focused on making the most Paperlike epaper display that has no ghosting and high refresh rate - 60 to 120fps. We started workin
15.
Vulnerabilities across keyboard apps reveal keystrokes to network eavesdroppers(citizenlab.ca)
4 points by up6w6 2y ago|0 comments
16.
Mitigating a token-length side-channel attack in our AI products(blog.cloudflare.com)
2 points by up6w6 2y ago|0 comments
17.
Linus Torvalds Isn't Happy with Some of the Bcachefs Code for Linux 6.9(phoronix.com)
5 points by up6w6 3y ago|2 comments
18.
by up6w6 3y ago
The Opus model that seems to perform better than GPT4 is unfortunately much more expensive than the OpenAI model. Pricing (input/output per million tokens): GPT4-turbo: $10/$30 Claude 3 Opus: $15/$75
19.
by up6w6 3y ago
Chatgpt-3.5 price reduction seems to be a direct response to Mixtral, which was cheaper (~0.0019 vs 0.0020 for 1K tokens) and better ( https://arena.lmsys.org/ ) until now.
20.
by up6w6 3y ago
Sadly we will still have to wait for application. Firefox doesn't have HDR support even on Windows right now.
21.
by up6w6 3y ago
I think the medium is trying to compete with Anthropic's Claude than Openai's products https://www-files.anthropic.com/production/images/model_pric...
22.
by up6w6 3y ago
For shell commands, there is a tool from Github that does it quite well (and it's included on the copilot plan)[1]. It's perfect to run fast bash commands that I forgot but are simple enough that the first Google search result wou
23.
by up6w6 3y ago
Love how they want to keep the focus to the personal usage instead of selling themselves into building features for business. Does anyone knows what is the current state of the voice commands? I always wanted a self hosted version of Google
24.
by up6w6 3y ago
Check out https://github.com/seemoo-lab/openhaystack
25.
Intel makes Thunderbolt 5 official, promising speeds of up to 120Gbps(theverge.com)
29 points by up6w6 3y ago|16 comments
26.
by up6w6 3y ago
True. The results from codex are actually from code-cushman-001 (Chen et al. 2021), which is an older model that Copilot was based on.
27.
by up6w6 3y ago
Even the 7B model of code llama seems to be competitive with Codex, the model behind copilot https://ai.meta.com/blog/code-llama-large-language-model-cod...
28.
by up6w6 3y ago
> usually can have their TDP modified in BIOS, via Ryzen Master, or third party tools like RyzenAdj (Phoenix support: https://github.com/FlyGoat/RyzenAdj/pull/256 ) to perform pretty closely. I didn't
29.
Kagi Raises $670K(blog.kagi.com)
9 points by up6w6 3y ago|3 comments
30.
by up6w6 3y ago
Usernames is a pending feature[1], something discussed in their "secret" non-indexed forum. [1] https://community.signalusers.org/t/usernames-in-signal/9157

More ›