From 9280becc49911c32f60f5f31b7cd5d6c97100a58 Mon Sep 17 00:00:00 2001 From: Paul Buetow Date: Sun, 31 May 2026 14:40:38 +0300 Subject: Update content for gemtext --- gemfeed/2026-06-01-gt-calculator.gmi | 2 +- gemfeed/2026-06-01-gt-calculator.gmi.tpl | 2 +- gemfeed/atom.xml | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'gemfeed') diff --git a/gemfeed/2026-06-01-gt-calculator.gmi b/gemfeed/2026-06-01-gt-calculator.gmi index c47c2a93..8d73c08b 100644 --- a/gemfeed/2026-06-01-gt-calculator.gmi +++ b/gemfeed/2026-06-01-gt-calculator.gmi @@ -490,7 +490,7 @@ For the complete and always-up-to-date feature guide, detailed docs for every fe => https://codeberg.org/snonux/gt gt on Codeberg -But will I now invest a couple of thousand dollars in hardware to run Qwen 2.5 35B or 27B? (I used the dense 27B model most of the time to build `gt`). Unfortunately, no. I don't think it's worth the cost yet, as cloud models are still cheaper and more convenient. +But will I now invest a couple of thousand dollars in hardware to run Qwen 2.5 35B or 27B? (I used the dense 27B model most of the time to build `gt`). Unfortunately, no. I don't think it's worth the cost yet, as cloud models are still cheaper and more convenient and the frontier ones also much more "intelligent". However, I will keep an eye on how the technology develops and continue experimenting with rented Hyperstack VMs for now; I will also default more often to smaller LLMs that could potentially run on home hardware. Ollama Cloud subscription or an OpenRouter API key are also good options alongside Claude and OpenAI Codex. diff --git a/gemfeed/2026-06-01-gt-calculator.gmi.tpl b/gemfeed/2026-06-01-gt-calculator.gmi.tpl index 2cba2842..2a684649 100644 --- a/gemfeed/2026-06-01-gt-calculator.gmi.tpl +++ b/gemfeed/2026-06-01-gt-calculator.gmi.tpl @@ -464,7 +464,7 @@ For the complete and always-up-to-date feature guide, detailed docs for every fe => https://codeberg.org/snonux/gt gt on Codeberg -But will I now invest a couple of thousand dollars in hardware to run Qwen 2.5 35B or 27B? (I used the dense 27B model most of the time to build `gt`). Unfortunately, no. I don't think it's worth the cost yet, as cloud models are still cheaper and more convenient. +But will I now invest a couple of thousand dollars in hardware to run Qwen 2.5 35B or 27B? (I used the dense 27B model most of the time to build `gt`). Unfortunately, no. I don't think it's worth the cost yet, as cloud models are still cheaper and more convenient and the frontier ones also much more "intelligent". However, I will keep an eye on how the technology develops and continue experimenting with rented Hyperstack VMs for now; I will also default more often to smaller LLMs that could potentially run on home hardware. Ollama Cloud subscription or an OpenRouter API key are also good options alongside Claude and OpenAI Codex. diff --git a/gemfeed/atom.xml b/gemfeed/atom.xml index 6e3566e3..99fb0a91 100644 --- a/gemfeed/atom.xml +++ b/gemfeed/atom.xml @@ -1,6 +1,6 @@ - 2026-05-31T14:37:53+03:00 + 2026-05-31T14:40:26+03:00 foo.zone feed To be in the .zone! @@ -607,7 +607,7 @@ http://www.gnu.org/software/src-highlite -->
gt on Codeberg

-But will I now invest a couple of thousand dollars in hardware to run Qwen 2.5 35B or 27B? (I used the dense 27B model most of the time to build gt). Unfortunately, no. I don't think it's worth the cost yet, as cloud models are still cheaper and more convenient.
+But will I now invest a couple of thousand dollars in hardware to run Qwen 2.5 35B or 27B? (I used the dense 27B model most of the time to build gt). Unfortunately, no. I don't think it's worth the cost yet, as cloud models are still cheaper and more convenient and the frontier ones also much more "intelligent".

However, I will keep an eye on how the technology develops and continue experimenting with rented Hyperstack VMs for now; I will also default more often to smaller LLMs that could potentially run on home hardware. Ollama Cloud subscription or an OpenRouter API key are also good options alongside Claude and OpenAI Codex.

-- cgit v1.2.3