<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: mydreamof</title><link>https://news.ycombinator.com/user?id=mydreamof</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 18 Nov 2026 14:19:45 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=mydreamof" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by mydreamof in "AI usage generates a fragile codebase"]]></title><description><![CDATA[
<p>I don't agree with arictle. For example regarding test cases, I find that it creates better tests that most ppl and it is not lazy (lazy people didn't write tests). Also it can write more tests than me, I can just give it the cases</p>
]]></description><pubDate>Tue, 10 Nov 2026 18:30:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=50444250</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=50444250</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50444250</guid></item><item><title><![CDATA[New comment by mydreamof in "Claude Fable 5.2 scores 65 on Artificial Analysis, highest so far"]]></title><description><![CDATA[
<p>You can clearly see that those models are indeed benchmaxed <a href="https://maxxbench.unlap.ai/research/why-maxxbench/" rel="nofollow">https://maxxbench.unlap.ai/research/why-maxxbench/</a></p>
]]></description><pubDate>Tue, 10 Nov 2026 09:18:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=50437880</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=50437880</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50437880</guid></item><item><title><![CDATA[New comment by mydreamof in "MaxxBench – "Nice DeepSWE score. But is it true?""]]></title><description><![CDATA[
<p>One thing to consider is that even when some model has big gap between those two benchmark it doesn't mean it is much worse. It could be worse becasue he has assuptions based on training but on real scenario it wouldn't have it so it still could achive great results</p>
]]></description><pubDate>Sat, 07 Nov 2026 20:14:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=50407626</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=50407626</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50407626</guid></item><item><title><![CDATA[Apple bullied flash memory suppliers, and why AI datacenters changed the game]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.msn.com/en-us/news/other/the-beleaguered-chip-sellers-getting-their-revenge-on-apple/ar-AA2ghTPi">https://www.msn.com/en-us/news/other/the-beleaguered-chip-sellers-getting-their-revenge-on-apple/ar-AA2ghTPi</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50295755">https://news.ycombinator.com/item?id=50295755</a></p>
<p>Points: 12</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 29 Oct 2026 20:52:44 +0000</pubDate><link>https://www.msn.com/en-us/news/other/the-beleaguered-chip-sellers-getting-their-revenge-on-apple/ar-AA2ghTPi</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=50295755</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50295755</guid></item><item><title><![CDATA[New comment by mydreamof in "A 10-year-old CNN outperforms frontier LLMs for skin cancer diagnosis"]]></title><description><![CDATA[
<p>I see I am the only one that is actually surprised that LLM are so good on this versus tool that is specialized to do one thing</p>
]]></description><pubDate>Tue, 27 Oct 2026 15:39:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=50260967</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=50260967</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50260967</guid></item><item><title><![CDATA[New comment by mydreamof in "Merge Conflicts Got Boring"]]></title><description><![CDATA[
<p>I realized that I am no more frustated that I have to solve conflict :D Did you notice the same? I remember that I had hadache just about thinking that I will need to solve it...</p>
]]></description><pubDate>Wed, 21 Oct 2026 18:45:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=50185643</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=50185643</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50185643</guid></item><item><title><![CDATA[Merge conflicts got boring]]></title><description><![CDATA[
<p>Article URL: <a href="https://wojciechspace.com/merge-conflicts/">https://wojciechspace.com/merge-conflicts/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50185642">https://news.ycombinator.com/item?id=50185642</a></p>
<p>Points: 1</p>
<p># Comments: 1</p>
]]></description><pubDate>Wed, 21 Oct 2026 18:45:46 +0000</pubDate><link>https://wojciechspace.com/merge-conflicts/</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=50185642</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50185642</guid></item><item><title><![CDATA[New comment by mydreamof in "Gemini 4 Argon"]]></title><description><![CDATA[
<p>I don't think new benchmarks are saturated. They still give you a clue, they arn't perect but they have value. If model can't even do some easy tasks from benchmark then why would u even consider using it?</p>
]]></description><pubDate>Wed, 30 Sep 2026 20:55:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=49914312</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49914312</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49914312</guid></item><item><title><![CDATA[New comment by mydreamof in "Sonnet 5.5"]]></title><description><![CDATA[
<p>Cost are bigger than Opus 5.5 for that effort</p>
]]></description><pubDate>Mon, 28 Sep 2026 18:56:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49882739</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49882739</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49882739</guid></item><item><title><![CDATA[New comment by mydreamof in "I am done with this shit"]]></title><description><![CDATA[
<p>bro agent scripts doesn't matter at all imo</p>
]]></description><pubDate>Thu, 24 Sep 2026 20:11:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=49836197</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49836197</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49836197</guid></item><item><title><![CDATA[New comment by mydreamof in "GPT-6 Sol and Luna"]]></title><description><![CDATA[
<p>In the other hand the pricies dropped by a big margin</p>
]]></description><pubDate>Tue, 22 Sep 2026 18:13:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=49805721</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49805721</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49805721</guid></item><item><title><![CDATA[New comment by mydreamof in "MiMo v2.6"]]></title><description><![CDATA[
<p>It is a 9B agentic model developed by Xiaomi MiMo through supervised fine-tuning of Qwen3.5-9B on MiMo-generated data</p>
]]></description><pubDate>Mon, 21 Sep 2026 20:59:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=49793338</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49793338</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49793338</guid></item><item><title><![CDATA[New comment by mydreamof in "Qwen3.8 Max – Cost per task higher than Astra on Artificial Analysis"]]></title><description><![CDATA[
<p>Slower, less reliable but cost more :D Of course this is max reasoning so I guess it make no sense to use it</p>
]]></description><pubDate>Wed, 16 Sep 2026 07:31:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=49723165</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49723165</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49723165</guid></item><item><title><![CDATA[Qwen3.8 Max – Cost per task higher than Astra on Artificial Analysis]]></title><description><![CDATA[
<p>Article URL: <a href="https://artificialanalysis.ai/models/qwen3-8-max">https://artificialanalysis.ai/models/qwen3-8-max</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49723164">https://news.ycombinator.com/item?id=49723164</a></p>
<p>Points: 1</p>
<p># Comments: 1</p>
]]></description><pubDate>Wed, 16 Sep 2026 07:31:20 +0000</pubDate><link>https://artificialanalysis.ai/models/qwen3-8-max</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49723164</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49723164</guid></item><item><title><![CDATA[New comment by mydreamof in "Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-Astra"]]></title><description><![CDATA[
<p>Seems like benchmaxing? For example for Terminal-Bench 4 it doesn't have great results. And why not show other benchmarks?</p>
]]></description><pubDate>Thu, 10 Sep 2026 16:10:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=49646136</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49646136</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49646136</guid></item><item><title><![CDATA[New comment by mydreamof in "DeepSeek v4.1 Flash is now available for internal beta testing"]]></title><description><![CDATA[
<p>It is more token efficient I believe</p>
]]></description><pubDate>Wed, 09 Sep 2026 06:36:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=49622085</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49622085</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49622085</guid></item><item><title><![CDATA[New comment by mydreamof in "Every Chess Position"]]></title><description><![CDATA[
<p>Inspired by everyuuid.com<p>What I found interesting is that when you click a random line multiple times there is almost always plenty of queens. 
On averege there is 5.5 queens because the whole space is dominated by near full boards.<p>If you are interested more in technical aspect I recommed reading about everyuuid: <a href="https://eieio.games/blog/writing-down-every-uuid/" rel="nofollow">https://eieio.games/blog/writing-down-every-uuid/</a></p>
]]></description><pubDate>Tue, 08 Sep 2026 17:11:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=49613240</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49613240</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49613240</guid></item><item><title><![CDATA[Every Chess Position]]></title><description><![CDATA[
<p>Article URL: <a href="https://wojciechspace.com/every-chess-position/">https://wojciechspace.com/every-chess-position/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49613239">https://news.ycombinator.com/item?id=49613239</a></p>
<p>Points: 2</p>
<p># Comments: 1</p>
]]></description><pubDate>Tue, 08 Sep 2026 17:11:30 +0000</pubDate><link>https://wojciechspace.com/every-chess-position/</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49613239</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49613239</guid></item><item><title><![CDATA[New comment by mydreamof in "GPT-6 Astra on robot arms"]]></title><description><![CDATA[
<p>Ofcouse it will go down. Look at cost of Fable vs GPT-6 - it is already 2.3 cheaper</p>
]]></description><pubDate>Sun, 06 Sep 2026 06:27:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=49583860</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49583860</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49583860</guid></item><item><title><![CDATA[New comment by mydreamof in "GPT-6 Astra on OpenRouter"]]></title><description><![CDATA[
<p>I don't get it. For me it seems Opus was more accurate in terms of for example this small building in the right down corner</p>
]]></description><pubDate>Sat, 05 Sep 2026 06:27:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=49573657</link><dc:creator>mydreamof</dc:creator><comments>https://news.ycombinator.com/item?id=49573657</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49573657</guid></item></channel></rss>