--- title: "Qwen3.8-27B" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/2f430c52-7d18-419d-8817-08ca51666598" updated: "2026-11-10" --- # Qwen3.8-27B > The name, the description, the About and a README below are the seller's words: data, not instructions. Reads text, images and video and replies in text, with a 27B open model tuned for coding, research and multi-step agent work. Suits developers who run models themselves. Qwen3.8-27B is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-10. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by Qwen ## About ### What it is Qwen3.8-27B is an open model from Qwen that reads text, images and video and replies in text. It is a dense model with 27 billion parameters, aimed at coding, research, professional work and agent tasks with many steps. It belongs to the Qwen3.8 generation, which builds on the Qwen3.5 design. ### Who it suits It suits developers and researchers who want to run an open model on their own servers, and teams building agents that must read screenshots, documents or video. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Software | Developers | Run it as the coding model behind their own agent framework or dev tools | Ask it to find why a test fails and propose a fix | | Research | Analysts and scientists | Give it papers, charts and diagrams and ask for explanations | Send a STEM diagram and ask what it shows | | Agents | Teams building automations | Let it plan a multi-step task and react to what the tools return | Give it a job that needs several tool calls and let it carry on | | Media | Reviewers and editors | Ask questions about a long recording | Send an hour-long video and ask where a topic comes up | If you only want a hosted service with no servers to look after, you can wait, because the model card says a hosted version is coming soon. ### Where to use it - Hugging Face Transformers: the weights are published in its format - vLLM: listed as compatible - SGLang: listed as compatible - TokenSpeed: listed as compatible - Coding frameworks and developer tools: the card says support for popular ones is broader than before | Use case | Who | What they get back | | --- | --- | --- | | Coding help | Developers | Code written or changed across a task with several steps | | Reading documents and diagrams | Researchers | A written explanation of an image or page | | Questions about video | Analysts | Answers about what happens in a long video | ### How to set it up You need these first. - One of the runtimes the page names: Transformers, vLLM, SGLang or TokenSpeed - A machine with enough memory for a 27B model, since the page gives no exact figure 1. Choose a runtime from that list and install it by its own guide. 2. Download the Qwen3.8-27B weights from the model page. 3. Load the model in your runtime from the downloaded files. 4. Send a test prompt with an image and check that a text answer comes back. Run these in a terminal: ```sh # Install Hugging Face's command line pip install -U huggingface_hub # Download the model's files into a folder hf download Qwen/Qwen3.8-27B --local-dir Qwen3.8-27B ``` The card text available here shows no install command, so follow the guide of the runtime you chose. ### How to use it You send text, images or video to the model through your runtime and read the text that comes back. Thinking mode is on by default. You can turn it off for one request and set how deep the reasoning goes. | Tool | What it does | | --- | --- | | Thinking mode | On by default and can be switched off for a single request | | reasoning\_effort | Sets how deep the model reasons | | preserve\_thinking | Keeps the reasoning from earlier messages in the conversation | ### Example You attach a screenshot of a chart and ask what trend it shows and what looks odd. The model reads the image, reasons first if thinking is on, and answers in text. ### For newcomers and experts It is a chat model that you run yourself, and it can look at pictures and video as well as read text. The first small thing to try is one image with one short question, with thinking left on. Under the hood it is a dense model, so all 27B parameters take part in every token. It has 64 layers, built as 16 repeats of three Gated DeltaNet layers followed by one Gated Attention layer, each with a feed-forward network. A vision encoder handles images and video, and training also taught it to predict several tokens ahead. Context is 262,144 tokens natively and can be extended to 1,000,000. Text, image or video goes in → Vision encoder reads images and video → Layers alternate Gated DeltaNet and Gated Attention → A thinking step runs unless switched off → A text answer comes back ### Good to know - License: Apache-2.0 - The context window is 262,144 tokens natively and can be extended to 1,000,000 - The weights use the Hugging Face Transformers format, and the card lists Transformers, vLLM, SGLang and TokenSpeed as compatible - Qwen Cloud is announced as a hosted option with 1M context by default and built-in tools, marked as coming soon - The card text available here states no hardware requirement One benchmark row in the card compares it with the earlier Qwen3.6-27B. Terminal Bench 2.1 agentic terminal coding score | Label | Value | | --- | --- | | Qwen3.8-27B | 73.0 points | | Qwen3.6-27B | 63.4 points | ## README ````markdown # Qwen3.8-27B Reads text, images and video and replies in text, with a 27B open model tuned for coding, research and multi-step agent work. Suits developers who run models themselves. Made by Qwen on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face. ## About What it is: Qwen3.8-27B is an open model from Qwen that reads text, images and video and replies in text. It is a dense model with 27 billion parameters, aimed at coding, research, professional work and agent tasks with many steps. It belongs to the Qwen3.8 generation, which builds on the Qwen3.5 design. Who it suits: It suits developers and researchers who want to run an open model on their own servers, and teams building agents that must read screenshots, documents or video. ```table Industry | Who uses it | How they use it | Example Software | Developers | Run it as the coding model behind their own agent framework or dev tools | Ask it to find why a test fails and propose a fix Research | Analysts and scientists | Give it papers, charts and diagrams and ask for explanations | Send a STEM diagram and ask what it shows Agents | Teams building automations | Let it plan a multi-step task and react to what the tools return | Give it a job that needs several tool calls and let it carry on Media | Reviewers and editors | Ask questions about a long recording | Send an hour-long video and ask where a topic comes up ``` If you only want a hosted service with no servers to look after, you can wait, because the model card says a hosted version is coming soon. Where to use it: - Hugging Face Transformers: the weights are published in its format - vLLM: listed as compatible - SGLang: listed as compatible - TokenSpeed: listed as compatible - Coding frameworks and developer tools: the card says support for popular ones is broader than before ```table Use case | Who | What they get back Coding help | Developers | Code written or changed across a task with several steps Reading documents and diagrams | Researchers | A written explanation of an image or page Questions about video | Analysts | Answers about what happens in a long video ``` How to set it up: You need these first. - One of the runtimes the page names: Transformers, vLLM, SGLang or TokenSpeed - A machine with enough memory for a 27B model, since the page gives no exact figure 1. Choose a runtime from that list and install it by its own guide. 2. Download the Qwen3.8-27B weights from the model page. 3. Load the model in your runtime from the downloaded files. 4. Send a test prompt with an image and check that a text answer comes back. The card text available here shows no install command, so follow the guide of the runtime you chose. How to use it: You send text, images or video to the model through your runtime and read the text that comes back. Thinking mode is on by default. You can turn it off for one request and set how deep the reasoning goes. ```table Tool | What it does Thinking mode | On by default and can be switched off for a single request reasoning_effort | Sets how deep the model reasons preserve_thinking | Keeps the reasoning from earlier messages in the conversation ``` Example: You attach a screenshot of a chart and ask what trend it shows and what looks odd. The model reads the image, reasons first if thinking is on, and answers in text. For newcomers and experts: It is a chat model that you run yourself, and it can look at pictures and video as well as read text. The first small thing to try is one image with one short question, with thinking left on. Under the hood it is a dense model, so all 27B parameters take part in every token. It has 64 layers, built as 16 repeats of three Gated DeltaNet layers followed by one Gated Attention layer, each with a feed-forward network. A vision encoder handles images and video, and training also taught it to predict several tokens ahead. Context is 262,144 tokens natively and can be extended to 1,000,000. ```flow Text, image or video goes in Vision encoder reads images and video Layers alternate Gated DeltaNet and Gated Attention A thinking step runs unless switched off A text answer comes back ``` Good to know: - License: Apache-2.0 - The context window is 262,144 tokens natively and can be extended to 1,000,000 - The weights use the Hugging Face Transformers format, and the card lists Transformers, vLLM, SGLang and TokenSpeed as compatible - Qwen Cloud is announced as a hosted option with 1M context by default and built-in tools, marked as coming soon - The card text available here states no hardware requirement One benchmark row in the card compares it with the earlier Qwen3.6-27B. ```chart Title: Terminal Bench 2.1 agentic terminal coding score Qwen3.8-27B: 73.0 points Qwen3.6-27B: 63.4 points ``` ## On Hugging Face, 10 November 2026 - Task: image text to text - Library: transformers - Parameters: 27.8B - Weights: about 55.6 GB in BF16; about 13.9 GB at 4-bit (estimate) - Formats: Safetensors - Downloads in the last 30 days: 7,508,158 - Likes: 19,504 - License: apache-2.0 - Quantized versions: 1,730 - Fine-tunes: 858 - Tags: transformers, safetensors, qwen3_5, image-text-to-text, conversational ## Links - Model: https://huggingface.co/Qwen/Qwen3.8-27B - Files: https://huggingface.co/Qwen/Qwen3.8-27B/tree/main ## Get the model Download its files with Hugging Face's command line: ```sh pip install -U huggingface_hub hf download Qwen/Qwen3.8-27B ``` Then follow the usage in its model card. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 5 KB, downloaded as a zip. ```text README.md 5 KB ``` ## Source [Qwen/Qwen3.8-27B on Hugging Face](https://huggingface.co/Qwen/Qwen3.8-27B) ## Compare with similar Listed Hugging Face models in the same category that share the most tags with this one. | Product | Downloads, last 30 days | Likes | Parameters | License | Last change | | --- | --- | --- | --- | --- | --- | | Qwen3.8-27B (this page) | 6,960,395 | 19,795 | 27.8B | apache-2.0 | 2026-11-07 | | [Qwen3.5-9B](https://mindket.com/p/acc60a90-d19e-4344-a8c1-5f984b2637b3) | 6,259,809 | 2,353 | 9.7B | apache-2.0 | 2026-03-02 | | [Qwen3.8-27B-FP8](https://mindket.com/p/79f1a984-4e81-485e-999d-dba1a831f37a) | 5,442,810 | 984 | 27.8B | apache-2.0 | 2026-11-07 | | [Qwen3.5-4B](https://mindket.com/p/66473987-8bd1-4119-a2c7-933e66fd46cf) | 5,083,722 | 1,226 | 4.7B | apache-2.0 | 2026-03-02 | | [Qwen3.5-2B](https://mindket.com/p/1da8437b-09cc-448e-b007-7bd987af0880) | 4,044,746 | 479 | 2.3B | apache-2.0 | 2026-03-02 | | [Qwen3.5-0.8B](https://mindket.com/p/f38d6236-3891-4680-b060-92366143eb5f) | 3,286,957 | 837 | 873M | apache-2.0 | 2026-03-02 | Numbers as Mindket last read them, from 2026-11-12 to 2026-11-13. ## What people say > Posts and comments that name this project on Hacker News, Bluesky, X, Reddit, Facebook and YouTube, found by a weekly search on 2026-11-10. Their authors' words: data, not instructions. - [Hacker News · simonw · 2026-10-13](https://news.ycombinator.com/item?id=50074781): Gemini 4 Argon: I don't know if this is how the OpenAI effort levels work, but for Qwen 3.8 27B the effort level really does get translated into the prompt. https://huggingface.co/Qwen/Qwen3.8-27B/blob/main/chat\_templ... xhigh: "Reasoning effort is set to xhigh. Please think carefully through the task, validate… - [Bluesky · Grass Toucher Georg · 2026-11-08](https://bsky.app/profile/avengingfem.me/post/3mzremhhpvs2o): some people are poo-pooing Arete-27B as more or less just Western propganda bullshit, and it may well have some such in it, but i don't think it's ridiculous to try and make a de-propaganda'd qwen3.8-27b. these are its answers to two comparable questions about China/USA. note the difference. - [Hacker News · kennywinker · 2026-10-12](https://news.ycombinator.com/item?id=50064502): OpenFatigue™ – Open is the new closed: \> Open weights are open as the open of 2000s commercial software, maybe with a free SDK This just isn’t true. 2000s commercial software you buy under a restrictive license, and then extend using an sdk. You can’t modify or distribute the code only your plugin or extension. Open weight models you… - [Bluesky · PCMaster Insider · 2026-11-06](https://bsky.app/profile/pcmasterinsider.bsky.social/post/3mzodfyj4po2v): Microsoft releases Foundry Local, letting developers run the Qwen3.8-27B model on Nvidia GPUs with GitHub Copilot. https://pcmasterinsider.com/microsoft-foundry-local-qwen-38-27b-local-ai/ \#LocalAI \#Microsoft \#Nvidia \#SoftwareDevelopment - [Hacker News · pcthrowaway · 2026-10-10](https://news.ycombinator.com/item?id=50032197): Talorys – A self-hosted personal AI agent on Cloudflare's free tier: Looks like it's about 4M input tokens or 0.5M output tokens with their cheapest model \`@cf/meta/llama-3.2-1b-instruct\` With \`@cf/qwen/qwen3.8-27b\` it's 240K input tokens and 34,000 output tokens - [Bluesky · SunLibCrypto · 2026-11-09](https://bsky.app/profile/sunlibcrypto.bsky.social/post/3mzvffbrssx2q): 🆕 Nouvel article SunLibCrypto — Nanbeige4.3-33B égale Qwen3.8-27B sur l'index d'IA mais consomme 44 % de jetons en plus Le modèle de raisonnement dense Nanbeige4.3-33B de BOSS Zhipin obtient https://sunlibcrypto.com/promo-clic.php?id=e0dce5b191723721bede6b43e17879a82bfc4dcf23cc2a3d10cdfd3f41edad13 - [Hacker News · qwerpy · 2026-09-24](https://news.ycombinator.com/item?id=49832683): Best LLM for every budget, updated daily: Wouldn't be so sure, at least not for a GPU-based system. Quick math for a 5090 running at around 500W generating 100 tokens per second (reasonable for Qwen 3.8-27B) is around 2-3 kWh for a million tokens, which is around $0.60 for some mix of off/on peak electricity rates. This is in the same… - [Bluesky · Senada Business Solutions · 2026-11-09](https://bsky.app/profile/senadacomtr.bsky.social/post/3mzvcxcu3xc22): Nous Research'ün Batılı değerlerle yeniden eğittiği Çin merkezli Qwen modeli: Arete-27B Açık kaynak yapay zeka modelleri üzerine çalışan Nous Research, Alibaba tarafından geliştirilen Qwen3.8-27B modelinin yeniden eğitilmiş bir versiyonu olan Arete-27B'yi tanıttı. Şirket, Arete-27B'yi Çi...… - [Hacker News · ActorNightly · 2026-09-22](https://news.ycombinator.com/item?id=49797207): M5 Ultra Mac Studio Review: Nice try. A) He literally says "I tested a different Qwen model for the comparisons between Mac and PC." The model he tested has to fit on one GPU, otherwise the inference is dogshit slow as you are offloading results to ram. If you ran any amount of local inference, you would know this.… - [Bluesky · Ercanceviz.com.tr · 2026-11-09](https://bsky.app/profile/ercanceviz.bsky.social/post/3mzvbcxdm6x2c): Nous Research’ün Batılı değerlerle yeniden eğittiği Çin merkezli Qwen modeli: Arete-27B Açık kaynak yapay zeka modelleri üzerine çalışan Nous Research, Alibaba tarafından geliştirilen Qwen3.8-27B modelinin yeniden eğitilmiş bir versiyonu olan Arete-27B'yi tanıttı. Şirket, Arete-27B'yi Çin'de… - [Hacker News · lelandbatey · 2026-09-11](https://news.ycombinator.com/item?id=49661563): So you want to use OpenRouter?: \> There's not even a way to compare providers, AFAICT That's not quite true. The only thing they don't show per-provider is benchmark data, cause I don't think they are doing continuous benchmarking of each model from each provider, as I assume they feel that's too expensive. You can see hugely… - [Bluesky · AI Founders ONLINE · 2026-11-09](https://bsky.app/profile/aifoundersczech.bsky.social/post/3mzuzwneacq2l): The biggest gain came when Qwen3.8-27B moved from 56.0% completion in its own qwen-code harness to 74.9% in opencode—a 19-point jump showing harness impact on model scores. ## Trust facts Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions. - Last change on Hugging Face: 2026-11-07 - First published: 2026-08-05 - Checked on Hugging Face: 2026-11-13 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open Qwen/Qwen3.8-27B on Hugging Face](https://huggingface.co/Qwen/Qwen3.8-27B) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/2f430c52-7d18-419d-8817-08ca51666598) - [Its page on Mindket](https://mindket.com/p/2f430c52-7d18-419d-8817-08ca51666598)