<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Apple Intelligence Foundation Language Models · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2407.21075/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/2407.21075/feed.xml" rel="self" type="application/rss+xml" /><item><title>Apple Intelligence Foundation Language Models — mentioned by dialup_sounds</title><link>https://news.ycombinator.com/item?id=44567264</link><guid isPermaLink="false">urn:hackerpapers:mention:2407.21075:44567264</guid><pubDate>Tue, 15 Jul 2025 02:18:39 GMT</pubDate><description>&lt;p&gt;https://arxiv.org/abs/2407.21075 tl;dr - they used Google TPUs&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2407.21075/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.AI</category></item><item><title>Apple Intelligence Foundation Language Models — mentioned by future10se</title><link>https://news.ycombinator.com/item?id=43714969</link><guid isPermaLink="false">urn:hackerpapers:mention:2407.21075:43714969</guid><pubDate>Thu, 17 Apr 2025 10:31:46 GMT</pubDate><description>&lt;p&gt;The on-device models used for Apple Intelligence (writing tools, notification and email/message summaries, etc.) are around ~3B parameters. I mean, they could be better (to put it nicely), but there is a legitimate use-case for them and I&amp;#x27;d love to see more work in this space. https://machinelearning.apple.com/research/introducing-apple... https://arxiv.org/abs/2407.21075&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2407.21075/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.AI</category></item><item><title>Apple Intelligence Foundation Language Models — mentioned by dialup_sounds</title><link>https://news.ycombinator.com/item?id=41412742</link><guid isPermaLink="false">urn:hackerpapers:mention:2407.21075:41412742</guid><pubDate>Sat, 31 Aug 2024 23:12:35 GMT</pubDate><description>&lt;p&gt;https://arxiv.org/abs/2407.21075 AFM-server was trained on 8192 TPUv4 chips Someone more versed can say if that is huge or not.&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2407.21075/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.AI</category></item><item><title>Apple Intelligence Foundation Language Models — HN submission</title><link>https://news.ycombinator.com/item?id=41204287</link><guid isPermaLink="false">urn:hackerpapers:submission:2407.21075:41204287</guid><pubDate>Fri, 09 Aug 2024 18:38:35 GMT</pubDate><description>&lt;p&gt;Apple Intelligence Foundation Language Models&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;56 points; 23 discussion replies on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/2407.21075/"&gt;Paper and source history&lt;/a&gt;. A submission is not an endorsement.&lt;/p&gt;</description><category>cs.AI</category></item></channel></rss>
