Deathmax

24 ms·

1.
by Deathmax 2mo ago
Except that officially, that is not what they do? It's a dick move not to put the two usecases under separate user agents, but their documentation says you're free to block Google-Extended via robots.txt which is used for training
2.
by Deathmax 2mo ago
The inverters are meant to come with anti-islanding to detect when it's been disconnected from the grid, which applies to both regular PV installs and a plug-in panel.
3.
by Deathmax 3mo ago
autokbisw solved the per-physical keyboard layout problem for me, switching between the Macbook's layout, a US keyboard, and a UK keyboard.
4.
by Deathmax 4mo ago
All new/renewing enterprise contracts with Claude Enterprise and ChatGPT Enterprise no longer offer usage-based subscriptions, but instead will charge API pricing for all tokens consumed, and as you've said, the subs are better de
5.
by Deathmax 5mo ago
There's prior art from Claude's own scheduled tasks' jitter: https://code.claude.com/docs/en/scheduled-tasks#jitter > Recurring tasks fire up to 10% of their period late, capped at 15 minutes. An
6.
by Deathmax 5mo ago
OpenRouter tells you if they submit with your user ID or anonymously if you hover over one of the icons on the provider, eg OpenAI has "OpenRouter submits API requests to this provider with an anonymous user ID.", Azure OpenAI on
7.
by Deathmax 5mo ago
The comments are all from the same author. OpenRouter recently started enforcing account-level regional restrictions for providers that enforce it (OpenAI, Anthropic, Google) - ie blocking accounts that look like they are being used by user
8.
by Deathmax 5mo ago
Microsoft has been taking steps to mitigate the leaked code signing certificate problem. On the driver side of things, new versions of Windows no longer trust the cross-signed certs, so you must submit your driver to Microsoft to validate a
9.
by Deathmax 9mo ago
NordVPN calls out when a location is virtual, so unless ipinfo is claiming they have virtual locations that are not labelled as such, they are at least transparent about it. They did document the physical server locations of their virtual l
10.
by Deathmax 9mo ago
The author was running a quantised version of GLM 4.5 _Air_, not the full fat version. API pricing for that is closer to $0.2/$1.1 at the top end from z.ai themselves, half the price from Novita/SiliconFlow.
11.
by Deathmax 10mo ago
You sort of can on Android, but it's a few steps: 1. Trigger Circle to Search with long holding the home button/bar 2. Select the image 3. Navigate to About this image on the Google search top bar all the way to the right - check
12.
by Deathmax 10mo ago
Depending on which shared GCP project you get assigned to, mine had a global 300 million tokens per minute quota that was being hit regularly.
13.
by Deathmax 10mo ago
Bad news then, they've bumped 3.0 Pro pricing to $2/$12 ($4/$18 at long context).
14.
by Deathmax 10mo ago
So the Bigscreen Beyond 1 and 2? They've optimised entirely for weight at ~100 grams.
15.
by Deathmax 11mo ago
I can finally tear down my custom DERP server that I was using to get higher throughput between two NAT'd clients.
16.
by Deathmax 11mo ago
Not anymore, especially after other routers like Vercel's AI Gateway and proxies from LLM providers like Fal, DeepInfra, and AtlasCloud didn't get the memo of enforcing BYOK for ID verification required models after GPT-5's r
17.
by Deathmax 11mo ago
As if the App Store had any sort of those guarantees. I know of people have been scammed via WebView wrappers that purported to be some benign app to pass app store review, which were then pointed at fake exchange websites afterwards. GitLa
18.
by Deathmax 1y ago
Victims would almost certainly have transferred funds to money mules, who would have then immediately broken digital audit trails to the ultimate destination by withdrawing as cash before passing it around.
19.
by Deathmax 1y ago
Riot documents the need to have IOMMU support enabled for Vanguard: https://support-valorant.riotgames.com/hc/en-us/articles/222...
20.
by Deathmax 1y ago
Vertex's offering of Gemini very much does implicit caching, and has always been the case [1]. The recent addition of applying implicit cache hit discounts also works on Vertex, as long as you don't use the `global` endpoint and h
21.
by Deathmax 1y ago
Gemini uses SentencePiece [1], and the proprietary Gemini models share the same tokenizer vocabulary as Gemma [2, 3, 4]. Out of the large proprietary western AI labs (OpenAI, Anthropic, Google), only Anthropic with Claude 3 and newer lack l
22.
by Deathmax 1y ago
It's a change to the CA rules that was passed in https://cabforum.org/2022/04/06/ballot-csc-13-update-to-subs... to align OV certificate requirements with the EV ones (that enforces the use of HSMs/
23.
by Deathmax 1y ago
Since April 2023 they support custom OIDC providers[1], and as of April 2024 that was extended to the free plan as well[2], so you can bring your own auth. [1]: https://tailscale.com/kb/1240/sso-custom-oidc [2]:
24.
by Deathmax 1y ago
https://www.minimaxi.com is their website for the Chinese parent company 上海稀宇科技有限公司, https://minimax.io is their international website for the Singapore based company Nanonoble Pte Ltd that handles operations outside
25.
by Deathmax 1y ago
I did read it, and I even went to their eval repo. > At the time of writing, there are two major versions available for GPT-4 and GPT-3.5 through OpenAI’s API, one snapshotted in March 2023 and another in June 2023. openaichat/gpt-3
26.
by Deathmax 1y ago
Your linked article is specifically comparing two different versioned snapshots of a model and not comparing the same model across time. You've also made the mistake of conflating what's served via API platforms which are meant to
27.
by Deathmax 1y ago
Also known as Claude 3.5 Sonnet V2 on AWS Bedrock and GCP Vertex AI
28.
by Deathmax 1y ago
It's not a 4B parameter model. The E4B variant is 7B parameters with 4B loaded into memory when using per-layer embedding cached to fast storage, and without vision or audio support.
29.
by Deathmax 1y ago
Gemini's free tier will absolutely use your inputs for training [1], same with Mistral's free tier [2]. Anthropic and OpenAI let's you opt into data collection for discounted prices or free tokens. [1]: https://ai.
30.
by Deathmax 1y ago
Thoughts used to be available in the Gemini/Vertex APIs when Gemini 2.0 Flash Thinking Experimental was initially introduced [1][2], and subsequently disabled to the public (I assume hidden behind a visibility flag) shortly after DeepS

More ›