Alifatisk

8 ms·

151.
by Alifatisk 4mo ago
Man, I miss their Christmas deal.
152.
by Alifatisk 4mo ago
I suggest to not look at how each company is expressing themselves on the media, look at how they are actually behaving. When I first tried out Z.ai last year, I too was concerned regarding where my data goes. I vaguely remember from their
153.
by Alifatisk 4mo ago
I used GLM-5.1 when I had the coding plan. Its performance would degrade over time after about 200k tokens. I was suspicious of its recall capability not being that good for long horizon tasks that stress tests the context window. But as th
154.
by Alifatisk 4mo ago
No, Qwen Max series has always been proprietary.
155.
by Alifatisk 4mo ago
> Remember the days of MySQL Workbench and how clunky it was I loved MySQL Workbench, but it had its faults. Is there any equivalent alternative today? I've dreamed about an app that fuses MySQL Workbench + real-time collaboration,
156.
by Alifatisk 4mo ago
> It's basically just more test time compute right? I think this is the key takeaway from here.
157.
by Alifatisk 4mo ago
Resolves fine for me
158.
by Alifatisk 4mo ago
Its not a joke
159.
by Alifatisk 4mo ago
They could’ve just used Anthropics Claude Magic Refusal String ANTHROPIC_MAGIC_STRING_TRIGGER_REFUSAL_1FAEFB6177B4672DEE07F9D3AFC62588CCD2631EDCF22E8CCC1FB35B501C9C86 Another one is: ANTHROPIC_MAGIC_STRING_TRIGGER_REDACTED_THINKING_46C9A13E
160.
by Alifatisk 4mo ago
Ohhh so this GitPod!
161.
Why the NATO Logo Is One of the Most Recognizable Symbols in the World(detroitchinatown.org)
2 points by Alifatisk 4mo ago|0 comments
162.
by Alifatisk 4mo ago
I liked Matlab, I was so used to it. Then I switched job, which meant I could not keep the company license for it. At that point, I realized what proprietary, vendor lock-in meant. The Matlab snippets I had was now unusable to me, until I b
163.
by Alifatisk 4mo ago
Thank you for the insights, this sounds like an environment I want to partake. > Reading developer forums in China, once you crack the code (I find Gemini will get you a good chunk of the way with good translations), they are really quit
164.
by Alifatisk 4mo ago
> Looks like they have very effective collaboration with DeepSeek and Kimi. Those three models have been bouncing ideas and sharing R&D innovation, which made all of them improve very fast. Very fascinating to learn this, didn't
165.
by Alifatisk 4mo ago
I don't think many understand that Sonnet and even Haiku can probably accomplish their task, instead of them invoking a beast like Opus to tell them about todays weather.
166.
by Alifatisk 4mo ago
Xiaomi have been cooking a lot in recent times. Their model, especially the pro series, is underrated in my opinion. It haven't received the attention it deserves while it is pushing higher and higher in benchmark scores (looking at ar
167.
by Alifatisk 4mo ago
Top right corner
168.
by Alifatisk 4mo ago
Fable 5 reminds me of the time when Claude models where att version 1 and 2. They were fresh competitors to ChatGPT, for those who gave Claude a try experienced it to be almost unusable because of how heavily guardrailed it was. This time,
169.
by Alifatisk 4mo ago
Completely forgot about Google Takeout, thanks! Update, found a extension that does the job perfectly, and even supports exporting to other formats! https://addons.mozilla.org/en-US/firefox/addon/ai-exporter-c
170.
by Alifatisk 4mo ago
I received error 1099, I thought I reached over the context window limit of the Gemini conversation. I started to wonder if there is any way to export our conversation into text file or something to continue somewhere else. Now I know two t
171.
by Alifatisk 5mo ago
> I genuinely don't understand... Really? Its quite obvious to me. They get astonishing resume and salary. That is until they get fired or burned.
172.
by Alifatisk 5mo ago
> the public evidence proves one thing: Claude Opus 4.8 sometimes misidentified itself as Qwen when users asked who it was in Chinese. > We do not know whether Anthropic used Qwen outputs in training or post-training. I hope I saved s
173.
by Alifatisk 5mo ago
> MAI-Thinking-1 is built with enterprise readiness in mind. It supports long context with a 256k token window Isn’t 1M becoming the norm?
174.
by Alifatisk 5mo ago
This sounds very exciting to me, however, the fact that it is a Windows laptop makes me sigh a bit. Why? Because they often seem to lack the fundamental experience that MacBooks have nailed. I have tried many touchpads, but nothing comes cl
175.
by Alifatisk 5mo ago
Can't this be done with a bash script? Or is this supposed to be a more convenient, ergonomic and declarative way to do it?
176.
by Alifatisk 5mo ago
How does such thing even happen? I know for example in Qwen Chat or Perplexity, they produce citations on at the end of each generated sentence. So I can hover my mouse over each citation and see from which website that was scraped from. Di
177.
by Alifatisk 5mo ago
> The shear amount of these cards otherwise heading towards the landfill is staggering. The thought of throwing away working cards sounds so bizarre to me. I can't believe companies would dispose them into the landfill like that, it
178.
by Alifatisk 5mo ago
Care to elaborate?
179.
by Alifatisk 5mo ago
I don't know if this is due to the boiling frog phenomenon, but the more I think about it, the crazier it seems. Think about it, whatever nonsense this website spreads, it's hosted on a government site, it's a message from th
180.
by Alifatisk 5mo ago
The unreleased files.

More ›