{"id":411,"date":"2026-06-29T15:03:55","date_gmt":"2026-06-29T11:33:55","guid":{"rendered":"https:\/\/haghiri75.com\/en\/?p=411"},"modified":"2026-06-29T15:03:55","modified_gmt":"2026-06-29T11:33:55","slug":"maralgpt-mythos-9b-just-released-and-this-is-why-im-proud-of-my-project","status":"publish","type":"post","link":"https:\/\/haghiri75.com\/en\/maralgpt-mythos-9b-just-released-and-this-is-why-im-proud-of-my-project\/","title":{"rendered":"MaralGPT &#8220;Mythos&#8221; 9B just released, and this is why I&#8217;m proud of my project."},"content":{"rendered":"<p>The other day I was watching a youtube video about models such as Fable and the host of podcast mentioned people from China, Russia or Iran may distil these models and make their own versions and called it\u00a0<em>dangerous.<\/em><\/p>\n<p>Honestly, what I found out &#8211; as an Iranian individual &#8211; is how Dario Amodei (and in general people like him) made this type of propaganda to keep their best models a huge money-making machine and nothing more. Well, you can&#8217;t stop us from using models which are already hosted on <a href=\"https:\/\/openrouter.ai\">OpenRouter<\/a>. You can&#8217;t keep us from using <a href=\"https:\/\/ollama.com\">Ollama<\/a> or similar tools. Can you?<\/p>\n<p>Well this is not the whole story. I was messing with Qwen 3.5 as soon as it got released and I found out how dynamic this model is in terms of context window size and also reasoning. But since internet connection was super expensive back then and our country was at war, I didn&#8217;t bother to work on the model that much.<\/p>\n<h2>Until &#8220;Fable&#8221; came out!<\/h2>\n<p>When this model came out, I saw how many people lost their crap over it. It was really a good model in any way possible and it could pass pretty much any test you had in mind.<\/p>\n<p>After a few days, the model just went offline for &#8220;non American&#8221; people, even the ones living in US borders (imagine Andrej Karpathy can&#8217;t use a production of his own!). Well this made a huge desparation world wide.<\/p>\n<p>GLM 5.2 became the next cool kid in town, but it still had a problem. It was too big to be self-hosted. So now what is the solution?<\/p>\n<h2>Maral Mythos 9B is!<\/h2>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full\" src=\"https:\/\/huggingface.co\/MaralGPT\/MaralGPT-Mythos-9B-2606\/resolve\/main\/maral-mythos-logo.png\" width=\"1024\" height=\"1024\" \/><\/p>\n<p>Well, I had a simple secret dataset, and every time a new SOTA model was available on Open Router or Cloudflare AI (or any other aggregator like them) I have it updated.<\/p>\n<p>The dataset, is around 500 million tokens of Claude 4.8 Opus and Claude 5 Fable data. It is data from GPT 5.5 and similar models. So it&#8217;s worth a try on Qwen 3.5, isn&#8217;t it?<\/p>\n<p>And after the fine tune, the benchmark looks like this:<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full\" src=\"https:\/\/huggingface.co\/MaralGPT\/MaralGPT-Mythos-9B-2606\/resolve\/main\/benchmark-1.jpg\" width=\"1744\" height=\"1232\" \/><\/p>\n<p>So, I have decided to release the model to the public because we need open source models to fight &#8220;personal decisions&#8221; which can affect lives of millions.<\/p>\n<h2>Links to the models<\/h2>\n<p>Original Model: <a href=\"https:\/\/huggingface.co\/MaralGPT\/MaralGPT-Mythos-9B-2606\">https:\/\/huggingface.co\/MaralGPT\/MaralGPT-Mythos-9B-2606<\/a><br \/>\nGGUF Models: <a href=\"https:\/\/huggingface.co\/MaralGPT\/MaralGPT-Mythos-9B-2606-GGUF\">https:\/\/huggingface.co\/MaralGPT\/MaralGPT-Mythos-9B-2606-GGUF<\/a><\/p>\n<p>If you can cooperate on further development or deployment of the model, you can stay in touch with me here or over haghiri75[at]gmail.com.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>The other day I was watching a youtube video about models such as Fable and the host of podcast mentioned people from China, Russia or Iran may distil these models and make their own versions and called it\u00a0dangerous. Honestly, what I found out &#8211; as an Iranian individual &#8211; is how Dario Amodei (and in &hellip; <\/p>\n<p class=\"link-more\"><a href=\"https:\/\/haghiri75.com\/en\/maralgpt-mythos-9b-just-released-and-this-is-why-im-proud-of-my-project\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;MaralGPT &#8220;Mythos&#8221; 9B just released, and this is why I&#8217;m proud of my project.&#8221;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_mi_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_newsletter_tier_id":0,"footnotes":"","jetpack_publicize_message":"","jetpack_is_tweetstorm":false,"jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","enabled":false}}},"categories":[2,4],"tags":[36,23,37,39,38,11,13,22,49],"jetpack_publicize_connections":[],"aioseo_notices":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"jetpack_shortlink":"https:\/\/wp.me\/p8BkKn-6D","jetpack-related-posts":[{"id":150,"url":"https:\/\/haghiri75.com\/en\/why-i-love-yolov5\/","url_meta":{"origin":411,"position":0},"title":"Why I love YOLOv5?","author":"prp-e","date":"December 12, 2021","format":false,"excerpt":"I am a big fan of Nicholas Renotte's channel on YouTube. I also love computer vision and its combination with deep learning. A few months ago, Nicholas posted this video, which is about YOLOv5. I usually am too lazy to watch videos which are longer than 15 minutes and I\u2026","rel":"","context":"In &quot;Computer Architecture &amp; Programming&quot;","block_context":{"text":"Computer Architecture &amp; Programming","link":"https:\/\/haghiri75.com\/en\/category\/computer\/"},"img":{"alt_text":"","src":"https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/splash-1024x677.jpg?resize=350%2C200","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/splash-1024x677.jpg?resize=350%2C200 1x, https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/splash-1024x677.jpg?resize=525%2C300 1.5x"},"classes":[]},{"id":16,"url":"https:\/\/haghiri75.com\/en\/microcontrollers-design-and-implementation-released\/","url_meta":{"origin":411,"position":1},"title":"Microcontrollers, Design and Implementation released!","author":"prp-e","date":"April 4, 2017","format":false,"excerpt":"It was about two years I started serious study on computer architecture. In these years, I learned a lot and I could simulate and implement a microprocessor, similar to real ones. In Summer 2016, I decided to share my experience with others. Then, I started writing this book. This book\u2026","rel":"","context":"In &quot;Computer Architecture &amp; Programming&quot;","block_context":{"text":"Computer Architecture &amp; Programming","link":"https:\/\/haghiri75.com\/en\/category\/computer\/"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":192,"url":"https:\/\/haghiri75.com\/en\/i-was-to-cheap-to-pay-10-a-month-for-copilot-so-i-made-my-own\/","url_meta":{"origin":411,"position":2},"title":"I was too cheap to pay $10 a month for copilot, so I made my own","author":"prp-e","date":"September 4, 2022","format":false,"excerpt":"In mid 2021, there was a revolution in coding. As a lazy programmer who always needed a fast and smart assistant, I was really happy to have Github Copilot in my arsenal of coding tools. So I was one of the early adapters of the whole idea of AI pair\u2026","rel":"","context":"In &quot;Computer Architecture &amp; Programming&quot;","block_context":{"text":"Computer Architecture &amp; Programming","link":"https:\/\/haghiri75.com\/en\/category\/computer\/"},"img":{"alt_text":"My own copilot!","src":"https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/Screen-Shot-2022-09-03-at-10.46.46-PM.png?resize=350%2C200","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/Screen-Shot-2022-09-03-at-10.46.46-PM.png?resize=350%2C200 1x, https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/Screen-Shot-2022-09-03-at-10.46.46-PM.png?resize=525%2C300 1.5x"},"classes":[]},{"id":300,"url":"https:\/\/haghiri75.com\/en\/nucleus-is-the-proof-that-small-is-the-new-big\/","url_meta":{"origin":411,"position":3},"title":"Nucleus is the proof that &#8220;Small is the new Big&#8221;","author":"prp-e","date":"January 13, 2024","format":false,"excerpt":"No matter what you heard, size matters. Specially in the world of AI models, having a smaller and more affordable model is the key to win the competition. This is why Microsoft even invested time, GPU and money on Phi project, which is a Small Language Model or SLM for\u2026","rel":"","context":"In &quot;Computer Architecture &amp; Programming&quot;","block_context":{"text":"Computer Architecture &amp; Programming","link":"https:\/\/haghiri75.com\/en\/category\/computer\/"},"img":{"alt_text":"","src":"https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/CEO.png?resize=350%2C200","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/CEO.png?resize=350%2C200 1x, https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/CEO.png?resize=525%2C300 1.5x, https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/CEO.png?resize=700%2C400 2x"},"classes":[]},{"id":286,"url":"https:\/\/haghiri75.com\/en\/maral-is-here-7-billion-parameters-bilingual-model-with-support-of-persian\/","url_meta":{"origin":411,"position":4},"title":"Maral is here, 7 billion parameters bilingual model with support of Persian!","author":"prp-e","date":"December 28, 2023","format":false,"excerpt":"If you read my previous post, you know how much I like open source AI material, and I even jokingly titled my BLOOM post I was too cheap to pay for GitHub's copilot! So making an open source model was always one of my goals of life. Also, in my\u2026","rel":"","context":"In &quot;Computer Architecture &amp; Programming&quot;","block_context":{"text":"Computer Architecture &amp; Programming","link":"https:\/\/haghiri75.com\/en\/category\/computer\/"},"img":{"alt_text":"","src":"https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/maral-featured-2.png?resize=350%2C200","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/maral-featured-2.png?resize=350%2C200 1x, https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/maral-featured-2.png?resize=525%2C300 1.5x, https:\/\/i0.wp.com\/haghiri75.com\/en\/wp-content\/uploads\/maral-featured-2.png?resize=700%2C400 2x"},"classes":[]},{"id":354,"url":"https:\/\/haghiri75.com\/en\/lets-build-metaverse-with-ai-what-we-have\/","url_meta":{"origin":411,"position":5},"title":"Let&#8217;s build Metaverse with AI: What we have?","author":"prp-e","date":"November 22, 2024","format":false,"excerpt":"In the previous post about building metaverse with AI (link), I discussed the generic points of view, what we need and all the stuff like that. In this post, I am going to discuss about AI models we have which can be helpful in order to build the metaverse using\u2026","rel":"","context":"In &quot;Projects&quot;","block_context":{"text":"Projects","link":"https:\/\/haghiri75.com\/en\/category\/projects\/"},"img":{"alt_text":"","src":"https:\/\/i0.wp.com\/mann-e-images.storage.c2.liara.space\/75117f84-c86f-49cc-b41a-4734b8e2db94_part_0001.png?resize=350%2C200&ssl=1","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/mann-e-images.storage.c2.liara.space\/75117f84-c86f-49cc-b41a-4734b8e2db94_part_0001.png?resize=350%2C200&ssl=1 1x, https:\/\/i0.wp.com\/mann-e-images.storage.c2.liara.space\/75117f84-c86f-49cc-b41a-4734b8e2db94_part_0001.png?resize=525%2C300&ssl=1 1.5x, https:\/\/i0.wp.com\/mann-e-images.storage.c2.liara.space\/75117f84-c86f-49cc-b41a-4734b8e2db94_part_0001.png?resize=700%2C400&ssl=1 2x, https:\/\/i0.wp.com\/mann-e-images.storage.c2.liara.space\/75117f84-c86f-49cc-b41a-4734b8e2db94_part_0001.png?resize=1050%2C600&ssl=1 3x, https:\/\/i0.wp.com\/mann-e-images.storage.c2.liara.space\/75117f84-c86f-49cc-b41a-4734b8e2db94_part_0001.png?resize=1400%2C800&ssl=1 4x"},"classes":[]}],"_links":{"self":[{"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/posts\/411"}],"collection":[{"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/comments?post=411"}],"version-history":[{"count":2,"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/posts\/411\/revisions"}],"predecessor-version":[{"id":413,"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/posts\/411\/revisions\/413"}],"wp:attachment":[{"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/media?parent=411"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/categories?post=411"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/haghiri75.com\/en\/wp-json\/wp\/v2\/tags?post=411"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}