--- title: Try not to waste your tokens | daily.dev description: A developer shares a week of personal experiments with AI coding tools, including running out of weekly tokens on Ollama while working on two GitHub projects... canonical: https://daily.dev/posts/try-not-to-waste-your-tokens-tc1o5uf85 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Try not to waste your tokens | daily.dev og:description: A developer shares a week of personal experiments with AI coding tools, including running out of weekly tokens on Ollama while working on two GitHub projects... og:url: https://daily.dev/posts/try-not-to-waste-your-tokens-tc1o5uf85 og:image: https://api.daily.dev/og/posts/Tc1O5uf85.png og:image:alt: Try not to waste your tokens og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Try not to waste your tokens **[Özkan Pakdil](https://daily.dev/sources/yhtxohtmgtdk1f5lkdkse)** · [@ozkanpakdil](https://daily.dev/ozkanpakdil) · 1 min read · 0 upvotes · 0 comments ## Summary A developer shares a week of personal experiments with AI coding tools, including running out of weekly tokens on Ollama while working on two GitHub projects (TrendCast and github-playwright-monitor), studying machine learning with Hugging Face models, and trying DeepSeek's harness alongside GitHub Copilot in VS Code. They note Copilot sometimes got stuck in loops that DeepSeek's harness avoided, tested a tool called opengsd.net for iterative vibe coding (warning it burns tokens fast), and compared token usage between GLM 5.3 flash and DeepSeek 4 flash, noting GLM is smarter but pricier in tokens. ## Content Could not finish my weekly tokens thanks #ollama These two I mainly worked on [https://github.com/ozkanpakdil/TrendCast](https://github.com/ozkanpakdil/TrendCast) [https://github.com/ozkanpakdil/github-playwright-monitor](https://github.com/ozkanpakdil/github-playwright-monitor) study some machine learning and using those huggingface models oh today I started using deepseek harness, first it looked like another coding client but I think it is little bit better because when I use #vscode github copilot, model was getting forever loops but some how DHS deepseek harnes did not have that, so AI client is important too, I thought it was happening because of model quality but it was not, everyday learning something new 🤓 Also this week I heavily tested [https://opengsd.net/](https://opengsd.net/) this gives more iterative way to do vibe coding :) I dont like this term but what else we are going to call it, anyway. But be careful about GSD it uses a lot of tokens. GLM 5.3 flash has been good and it eats more token then deepseek 4 flash although smarter.  ## Similar posts on daily.dev - [Token Optimization: The Overlooked Skill for AI-Powered Developers](https://daily.dev/posts/token-optimization-the-overlooked-skill-for-ai-powered-developers-lnrugosqm) · Medium · 1 upvotes · 0 comments --- Tags: [#github](https://daily.dev/tags/github), [#ai-agents](https://daily.dev/tags/ai-agents), [#vibe-coding](https://daily.dev/tags/vibe-coding), [#ollama](https://daily.dev/tags/ollama), [#deepseek](https://daily.dev/tags/deepseek) [View this post on daily.dev](https://daily.dev/posts/try-not-to-waste-your-tokens-tc1o5uf85) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"DiscussionForumPosting","mainEntityOfPage":"https://daily.dev/posts/try-not-to-waste-your-tokens-tc1o5uf85","headline":"Try not to waste your tokens","text":"A developer shares a week of personal experiments with AI coding tools, including running out of weekly tokens on Ollama while working on two GitHub projects (TrendCast and github-playwright-monitor), studying machine learning with Hugging Face models, and trying DeepSeek's harness alongside GitHub Copilot in VS Code. They note Copilot sometimes got stuck in loops that DeepSeek's harness avoided, tested a tool called opengsd.net for iterative vibe coding (warning it burns tokens fast), and compared token usage between GLM 5.3 flash and DeepSeek 4 flash, noting GLM is smarter but pricier in tokens.","url":"https://daily.dev/posts/try-not-to-waste-your-tokens-tc1o5uf85","datePublished":"2026-08-30T22:02:28.071Z","dateModified":"2026-08-30T22:02:42.504Z","author":{"@type":"Person","name":"Özkan Pakdil","url":"https://daily.dev/ozkanpakdil","image":"https://media.daily.dev/image/upload/s--ihPlqRre--/f_auto/v1758259464/avatars/avatar_yHtxohtmgtdK1F5lKdKSe?_a=BAMAK+ZW0","description":"Coding since Commodore 64(1993), works with Linux/cloud/Java/spring boot/hibernate/C#/PostgreSQL","interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"EndorseAction"},"userInteractionCount":5860}},"interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"isPartOf":{"@type":"WebPage","url":"https://daily.dev/sources/yhtxohtmgtdk1f5lkdkse","name":"Özkan Pakdil"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Özkan Pakdil","item":"https://daily.dev/sources/yhtxohtmgtdk1f5lkdkse"},{"@type":"ListItem","position":3,"name":"Try not to waste your tokens"}]} ```