<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: Bolwin</title><link>https://news.ycombinator.com/user?id=Bolwin</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 18 Nov 2026 14:15:41 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=Bolwin" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by Bolwin in "Solid 2.0"]]></title><description><![CDATA[
<p>Try C# as a middle ground. Compiles faster than Rust, runs faster than Go</p>
]]></description><pubDate>Tue, 17 Nov 2026 21:32:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=50538489</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50538489</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50538489</guid></item><item><title><![CDATA[The Four Tiers of AI Intelligence]]></title><description><![CDATA[
<p>Article URL: <a href="https://blog.coship.fyi/blog/ai-tiers/">https://blog.coship.fyi/blog/ai-tiers/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50448477">https://news.ycombinator.com/item?id=50448477</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 11 Nov 2026 00:56:38 +0000</pubDate><link>https://blog.coship.fyi/blog/ai-tiers/</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50448477</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50448477</guid></item><item><title><![CDATA[New comment by Bolwin in "Claude Fable 5.2 and Mythos 5.2"]]></title><description><![CDATA[
<p>There is plenty of explanation, if you actually go seek out the benchmark. Past 80 benchmarks start approaching saturation and are not very useful to show differences</p>
]]></description><pubDate>Mon, 09 Nov 2026 20:54:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=50431982</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50431982</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50431982</guid></item><item><title><![CDATA[New comment by Bolwin in "Claude Fable 5.2 and Mythos 5.2"]]></title><description><![CDATA[
<p>Stop looking at fable max. On medium it's a bit better than Sol at $3.80. High and xhigh sit in between. In general you can ignore Max reasoning on Claude models if you care about price at all</p>
]]></description><pubDate>Mon, 09 Nov 2026 20:51:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=50431929</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50431929</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50431929</guid></item><item><title><![CDATA[New comment by Bolwin in "MaxxBench – "Nice DeepSWE score. But is it true?""]]></title><description><![CDATA[
<p>The benchmark makers and anyone who relies on benchmarks</p>
]]></description><pubDate>Sat, 07 Nov 2026 21:07:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=50408071</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50408071</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50408071</guid></item><item><title><![CDATA[New comment by Bolwin in "US Democrats Win Enough Seats Nationwide to End Presidential Electoral College"]]></title><description><![CDATA[
<p>I prefer approval voting. It's easier for the voter to understand and do and eliminates a form of spoiler effect ranked choice voting still has <a href="https://en.wikipedia.org/wiki/Approval_voting" rel="nofollow">https://en.wikipedia.org/wiki/Approval_voting</a></p>
]]></description><pubDate>Fri, 06 Nov 2026 19:10:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=50396892</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50396892</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50396892</guid></item><item><title><![CDATA[New comment by Bolwin in "Can Democrats effectively abolish Electoral College after midterm wins?"]]></title><description><![CDATA[
<p>Even if it is more representative?<p>The purpose of democracy is not to balance between arbitrary parties.</p>
]]></description><pubDate>Fri, 06 Nov 2026 18:54:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=50396672</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50396672</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50396672</guid></item><item><title><![CDATA[New comment by Bolwin in "Inkling 2 Small"]]></title><description><![CDATA[
<p>AA lists it as $0.30/0.03/1 and cost per task is $1.18, about double that of deepseek</p>
]]></description><pubDate>Fri, 06 Nov 2026 00:56:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=50386807</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50386807</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50386807</guid></item><item><title><![CDATA[New comment by Bolwin in "Artificial Analysis Intelligence Index v5"]]></title><description><![CDATA[
<p>Redacted is a bit extreme. They just haven't benched them yet</p>
]]></description><pubDate>Fri, 30 Oct 2026 22:02:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=50310731</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50310731</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50310731</guid></item><item><title><![CDATA[New comment by Bolwin in "Software may be getting worse, but it used to be horrible"]]></title><description><![CDATA[
<p>Interesting you should say this, as Microsoft just yesterday announced an update fixing this (gradual rollout).</p>
]]></description><pubDate>Fri, 30 Oct 2026 17:48:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=50307555</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50307555</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50307555</guid></item><item><title><![CDATA[New comment by Bolwin in "We won't "just hang out""]]></title><description><![CDATA[
<p>The joy of learning something new is very much tied to said thing being meaningful or useful. You could spend the rest of your life memorizing digits of pi and you'd be spending all of it "learning something new" but you wouldn't enjoy it very much.</p>
]]></description><pubDate>Fri, 30 Oct 2026 17:16:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=50307012</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50307012</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50307012</guid></item><item><title><![CDATA[New comment by Bolwin in "Kimi K3.1 Pro and K3.1 Pro Highspeed"]]></title><description><![CDATA[
<p>Kimi has sound, it's just very quiet</p>
]]></description><pubDate>Sun, 25 Oct 2026 00:07:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=50226807</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50226807</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50226807</guid></item><item><title><![CDATA[New comment by Bolwin in "Kimi K3.1 Pro and K3.1 Pro Highspeed"]]></title><description><![CDATA[
<p>The original k3 also matched its respective opus</p>
]]></description><pubDate>Fri, 23 Oct 2026 22:13:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=50216084</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50216084</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50216084</guid></item><item><title><![CDATA[New comment by Bolwin in "Kimi K3.1 scores 55 on Artificial Analysis, placed on the Pareto line"]]></title><description><![CDATA[
<p>This seems more a measure of anti promp injection than general intelligence</p>
]]></description><pubDate>Fri, 23 Oct 2026 21:56:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=50215919</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50215919</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50215919</guid></item><item><title><![CDATA[New comment by Bolwin in "Advanced coding in low-resource programming languages"]]></title><description><![CDATA[
<p>I hope someone runs multilangbench for other models. I really want to see how open models in particular do on it. Benchmaxxing by focusing on the top 5 languages seems an easy</p>
]]></description><pubDate>Wed, 21 Oct 2026 18:45:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=50185629</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50185629</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50185629</guid></item><item><title><![CDATA[New comment by Bolwin in "Voxtral 3"]]></title><description><![CDATA[
<p>That's fascinating. Are these for your keyboard?<p>You may have gone too fast honestly. Something like parakeet runs fast enough and seems to have much lower wer</p>
]]></description><pubDate>Tue, 20 Oct 2026 17:27:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=50169757</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50169757</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50169757</guid></item><item><title><![CDATA[New comment by Bolwin in "MiniMax M3.1 Flash"]]></title><description><![CDATA[
<p>Which claude model are you comparing it to? If you're using fable max obv it's going to be slower, but like sonnet high is both faster and more intelligent</p>
]]></description><pubDate>Tue, 20 Oct 2026 05:19:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=50161644</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50161644</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50161644</guid></item><item><title><![CDATA[New comment by Bolwin in "Frontier Pre-Training at Microsoft AI"]]></title><description><![CDATA[
<p>Very similar to Kimi K3 it seems, but fewer active params (80b vs 104b). I don't think they mentioned what form of attention they use</p>
]]></description><pubDate>Thu, 15 Oct 2026 19:31:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=50109014</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50109014</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50109014</guid></item><item><title><![CDATA[New comment by Bolwin in "GLM-Legal"]]></title><description><![CDATA[
<p>This is why I severely dislike using all pass rate. A model could be at like 93%, meeting practically every requirement except the broken ones and all pass would give it like 20%.</p>
]]></description><pubDate>Mon, 12 Oct 2026 17:44:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=50060006</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=50060006</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50060006</guid></item><item><title><![CDATA[New comment by Bolwin in "Claude Haiku 5.5"]]></title><description><![CDATA[
<p>I use it for title generation basically. Will have to see where this one fits in</p>
]]></description><pubDate>Wed, 07 Oct 2026 19:04:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=49997347</link><dc:creator>Bolwin</dc:creator><comments>https://news.ycombinator.com/item?id=49997347</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49997347</guid></item></channel></rss>