<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[Gemini 4.0 Argon下放到Google AI Pro方案了]]></title><description><![CDATA[<p dir="auto">我本來就有訂閱Gemini方案<br />
其實Gemini本身的訓練是文科生，語言能力跟輸出速度在業界是頂尖的<br />
但你千萬別用網頁或是App對話，他Harness的限制讓他非常的偷懶<br />
其實Antigrivity這個Agent蠻好用的，就他媽的只能接Gemini這個模型，有額度限制，<br />
像我20鎂/月方案一周大概不用20小時就燒光</p>
<p dir="auto">上個月的Gemini 3.8 Flash就讓我驚豔，目測輸出就是200~300Token/s，正確性也相當不錯<br />
因為我本身就有儲值DeepSeek我知道差距<br />
但是3.8 Flash最大的問題就是過度CoT，這也是類似老特講的把Harness綁進模型，<br />
常常一個要求他給我跑幾十步才給我結果，有時候搞得沒啥耐心<br />
我本人平常在開發時是完全不開思考模式的，這個無法關，就算思考開低也是給我幾十步。</p>
<p dir="auto">昨晚突然發現4.0 Argon下放到Pro方案<br />
我玩了三個半小時，生平第一次這麼久沒有對LLM飆過任何髒話的，沒出過bug或進度回頭，實作進度非常愉快<br />
但是天真的我在那三小時就把周額度就用光了<br />
<img src="https://upload.lcz.me/uploads/3ab653d8-9bca-4b42-8bf7-95c8da53d335.jpeg" alt="image.jpeg" class="img-markdown  img-fluid" /></p>
<p dir="auto">這個燒錢我用不起.... 掰<br />
我連3060都買不起了還搞這種貴族方案</p>
<p dir="auto">然後這個公開題目我一定測，實作的結果非常頂，我認為他早就背答案了，以後大家測試模型別拿這種東西來測了<br />
<img src="https://upload.lcz.me/uploads/66cd129c-1d17-4dc4-a0ba-2141420d4aaf.jpeg" alt="image.jpeg" class="img-markdown  img-fluid" /></p>
]]></description><link>https://lcz.me/topic/2571</link><generator>RSS for Node</generator><lastBuildDate>Sun, 25 Oct 2026 18:24:03 GMT</lastBuildDate><atom:link href="https://lcz.me/topic/2571.rss" rel="self" type="application/rss+xml"/><pubDate>Sun, 25 Oct 2026 07:08:33 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to Gemini 4.0 Argon下放到Google AI Pro方案了 on Sun, 25 Oct 2026 07:42:32 GMT]]></title><description><![CDATA[<blockquote>
<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/dardeaw-feng" aria-label="Profile: dardeaw-feng">@<bdi>dardeaw-feng</bdi></a> <a href="/post/27639">说</a>:</p>
<p dir="auto">antigrivity</p>
</blockquote>
<p dir="auto">装过 antigrivity ， 觉得不好用删了， 看来现在得装回来试试</p>
]]></description><link>https://lcz.me/post/27641</link><guid isPermaLink="true">https://lcz.me/post/27641</guid><dc:creator><![CDATA[johnnybegood]]></dc:creator><pubDate>Sun, 25 Oct 2026 07:42:32 GMT</pubDate></item><item><title><![CDATA[Reply to Gemini 4.0 Argon下放到Google AI Pro方案了 on Sun, 25 Oct 2026 07:39:23 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/johnnybegood" aria-label="Profile: johnnybegood">@<bdi>johnnybegood</bdi></a><br />
可以<br />
用AGY轉接進DSH<br />
但我自己測過原生的AGY效果比較好</p>
]]></description><link>https://lcz.me/post/27640</link><guid isPermaLink="true">https://lcz.me/post/27640</guid><dc:creator><![CDATA[fanweng-code]]></dc:creator><pubDate>Sun, 25 Oct 2026 07:39:23 GMT</pubDate></item><item><title><![CDATA[Reply to Gemini 4.0 Argon下放到Google AI Pro方案了 on Sun, 25 Oct 2026 07:38:26 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/johnnybegood" aria-label="Profile: johnnybegood">@<bdi>johnnybegood</bdi></a><br />
訂閱方案不給API, 請乖乖去買token<br />
但有antigrivity你也不需要去用dsh吧 這種生態綁定的就這樣</p>
<p dir="auto">其實dsh我裝好擱著的原因就是沒原生undo<br />
還要去找undo外掛我弄得也懶了<br />
那個fork功能好不習慣<br />
而antigrivity的undo是做到連代碼都能版控一起退回</p>
]]></description><link>https://lcz.me/post/27639</link><guid isPermaLink="true">https://lcz.me/post/27639</guid><dc:creator><![CDATA[dardeaw feng]]></dc:creator><pubDate>Sun, 25 Oct 2026 07:38:26 GMT</pubDate></item><item><title><![CDATA[Reply to Gemini 4.0 Argon下放到Google AI Pro方案了 on Sun, 25 Oct 2026 07:34:30 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/johnnybegood" aria-label="Profile: johnnybegood">@<bdi>johnnybegood</bdi></a> <a href="https://lcz.me/topic/1255">https://lcz.me/topic/1255</a> 不知道這些開源進化了沒？</p>
]]></description><link>https://lcz.me/post/27637</link><guid isPermaLink="true">https://lcz.me/post/27637</guid><dc:creator><![CDATA[David Chen]]></dc:creator><pubDate>Sun, 25 Oct 2026 07:34:30 GMT</pubDate></item><item><title><![CDATA[Reply to Gemini 4.0 Argon下放到Google AI Pro方案了 on Sun, 25 Oct 2026 07:33:39 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/david-chen" aria-label="Profile: David-Chen">@<bdi>David-Chen</bdi></a> 下次題目改 請動畫銓釋樹懶怎麼叫</p>
]]></description><link>https://lcz.me/post/27635</link><guid isPermaLink="true">https://lcz.me/post/27635</guid><dc:creator><![CDATA[dardeaw feng]]></dc:creator><pubDate>Sun, 25 Oct 2026 07:33:39 GMT</pubDate></item><item><title><![CDATA[Reply to Gemini 4.0 Argon下放到Google AI Pro方案了 on Sun, 25 Oct 2026 07:28:10 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/dardeaw-feng" aria-label="Profile: dardeaw-feng">@<bdi>dardeaw-feng</bdi></a> 以後改做『樹懶騎獨輪車，後面有隻獵豹在追她』，看看能不能畫出東西</p>
]]></description><link>https://lcz.me/post/27634</link><guid isPermaLink="true">https://lcz.me/post/27634</guid><dc:creator><![CDATA[David Chen]]></dc:creator><pubDate>Sun, 25 Oct 2026 07:28:10 GMT</pubDate></item><item><title><![CDATA[Reply to Gemini 4.0 Argon下放到Google AI Pro方案了 on Sun, 25 Oct 2026 07:26:30 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/dardeaw-feng" aria-label="Profile: dardeaw-feng">@<bdi>dardeaw-feng</bdi></a> 我也有 google AI pro ， argon 能用API接出来到 deepseek harness 么？</p>
]]></description><link>https://lcz.me/post/27633</link><guid isPermaLink="true">https://lcz.me/post/27633</guid><dc:creator><![CDATA[johnnybegood]]></dc:creator><pubDate>Sun, 25 Oct 2026 07:26:30 GMT</pubDate></item></channel></rss>