--- title: "Qwen3.8-27B GGUF (ShapeLearn Quantized)" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/ca45abfc-d3db-4fb5-aad9-dcf7a7a498c4" updated: "2026-11-12" --- # Qwen3.8-27B GGUF (ShapeLearn Quantized) > The name, the description, the About and a README below are the seller's words: data, not instructions. Runs Qwen3.8-27B as GGUF files quantized by ByteShape's ShapeLearn, in five GPU-sized options with vision input and MTP speculative decoding in llama.cpp. For people running it on their own GPU. Qwen3.8-27B GGUF (ShapeLearn Quantized) is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-12. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by byteshape ## About ### What it is This repository holds GGUF copies of Qwen3.8-27B quantized with ByteShape's ShapeLearn. ShapeLearn learns the best datatype for each tensor so quality holds up even at low bit lengths. The release replaces the earlier ShapeLearn-Lite versions, and each of its five models was scored against the BF16 baseline. ### Who it suits It suits people who run Qwen3.8-27B on their own GPU through llama.cpp and want to pick a size and quality trade-off that fits their card. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Software | Developers with a consumer GPU | Pick the model size that fits their GPU memory | Choose a smaller file for a card with less memory | | Research | People comparing quantizations | Read the quality-versus-speed chart on the card | Compare ShapeLearn files with other public GGUF copies | | Vision work | Users who send images | Run the model with the vision projector | Ask a question about a picture | People who want the unquantized weights can skip it, since every file here is quantized. ### Where to use it - llama.cpp, through llama-server - The web interface that llama-server serves on your own machine - GPUs such as the RTX 5090, which the card benchmarks, along with the 4090, 4080, 5060 Ti, 3090 and RTX PRO 6000 Blackwell - Windows, macOS or Linux, where you can download the files with Python | Use case | Who | What they get back | | --- | --- | --- | | Local chat and coding | Developers | A running llama-server with the model chosen by its Model Tag | | Vision questions | Users with images | Answers that use the auto-downloaded vision projector | | Faster decoding | Speed-minded users | MTP speculative decoding with no separate draft model | ### How to set it up You need these first: - llama.cpp, with build b10658 or newer if you want DFlash 2 - A GPU with room for the model you choose, from about 8.8 GB for the smallest of the five - Python and the huggingface\_hub package, if you want to download the files yourself 1. Pick a model from the card's table and copy its Model Tag. 2. Make sure your llama.cpp build is new enough. 3. Run llama-server with the tag, as shown below. ```sh llama-server -hf --mmproj-auto --spec-type draft-mtp --spec-draft-n-max 3 ``` ```sh llama-server --help | grep spec-type ``` ```sh pip install -U huggingface_hub hf download byteshape/Qwen3.8-27B-GGUF ``` ### How to use it Run llama-server with a Model Tag and open the web interface it serves on your machine. On the first run, llama.cpp fetches the model and the vision projector. The draft-mtp option turns on MTP speculative decoding, and the draft count option sets how many tokens are drafted per step. | Tool | What it does | | --- | --- | | --mmproj-auto | Loads the vision projector so the model can read images | | --spec-type draft-mtp | Turns on MTP speculative decoding using the head inside the GGUF | | --spec-draft-n-max | Sets how many tokens are drafted per step | ### Example You copy the Model Tag of one of the five models and run llama-server with it. llama.cpp downloads the model and the vision projector, then you open the web interface and chat. The card says speculative decoding adds another 1.4 to 1.8 times on top of plain decoding. ### For newcomers and experts Quantizing shrinks a model so it fits a smaller GPU, at some cost in quality. A first small step is to pick the smallest model in the table and run the command above. ShapeLearn picks a datatype per tensor, so each file mixes quantization types instead of using one for all. The card plots quality against tokens per second on an RTX 5090, with quality as the average score relative to BF16 on its benchmark suite. The speed figure is single-stream decoding with plain next-token prediction. The MTP head sits inside the GGUF, so you need no separate draft model. Pick a model from the table → Copy its Model Tag → Run llama-server with the tag → llama.cpp fetches model and projector → Chat in the web interface ### Good to know - Size: 27.3B parameters per Hugging Face, about 13.7 GB at 4-bit as an estimate, and the card's own files start at 8.8 GB - Evaluation: scored against BF16 on tests of math, code, knowledge, following instructions and tool use, in thinking and instruct modes - Versions: Hugging Face lists 1 quantized version made by other people - DFlash 2: it needs a llama.cpp build of b10658 or newer, and draft-mtp works on any build from that release on Model size in GB, first three of five models | Label | Value | | --- | --- | | GPU-1 | 8.8 GB | | GPU-2 | 9.9 GB | | GPU-3 | 10.4 GB | - License: apache-2.0 - Cost to run: none named, no API key or account - Data it reaches: on the first run llama.cpp fetches the model and its vision projector, and the web interface stays on your own machine - Not for: DFlash 2 on llama.cpp builds older than b10658, since that release added the support ## README ````markdown # Qwen3.8-27B GGUF (ShapeLearn Quantized) Runs Qwen3.8-27B as GGUF files quantized by ByteShape's ShapeLearn, in five GPU-sized options with vision input and MTP speculative decoding in llama.cpp. For people running it on their own GPU. Made by byteshape 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: This repository holds GGUF copies of Qwen3.8-27B quantized with ByteShape's ShapeLearn. ShapeLearn learns the best datatype for each tensor so quality holds up even at low bit lengths. The release replaces the earlier ShapeLearn-Lite versions, and each of its five models was scored against the BF16 baseline. Who it suits: It suits people who run Qwen3.8-27B on their own GPU through llama.cpp and want to pick a size and quality trade-off that fits their card. ```table Industry | Who uses it | How they use it | Example Software | Developers with a consumer GPU | Pick the model size that fits their GPU memory | Choose a smaller file for a card with less memory Research | People comparing quantizations | Read the quality-versus-speed chart on the card | Compare ShapeLearn files with other public GGUF copies Vision work | Users who send images | Run the model with the vision projector | Ask a question about a picture ``` People who want the unquantized weights can skip it, since every file here is quantized. Where to use it: - llama.cpp, through llama-server - The web interface that llama-server serves on your own machine - GPUs such as the RTX 5090, which the card benchmarks, along with the 4090, 4080, 5060 Ti, 3090 and RTX PRO 6000 Blackwell - Windows, macOS or Linux, where you can download the files with Python ```table Use case | Who | What they get back Local chat and coding | Developers | A running llama-server with the model chosen by its Model Tag Vision questions | Users with images | Answers that use the auto-downloaded vision projector Faster decoding | Speed-minded users | MTP speculative decoding with no separate draft model ``` How to set it up: You need these first: - llama.cpp, with build b10658 or newer if you want DFlash 2 - A GPU with room for the model you choose, from about 8.8 GB for the smallest of the five - Python and the huggingface_hub package, if you want to download the files yourself 1. Pick a model from the card's table and copy its Model Tag. 2. Make sure your llama.cpp build is new enough. 3. Run llama-server with the tag, as shown below. ```sh llama-server -hf --mmproj-auto --spec-type draft-mtp --spec-draft-n-max 3 ``` ```sh llama-server --help | grep spec-type ``` ```sh pip install -U huggingface_hub hf download byteshape/Qwen3.8-27B-GGUF ``` How to use it: Run llama-server with a Model Tag and open the web interface it serves on your machine. On the first run, llama.cpp fetches the model and the vision projector. The draft-mtp option turns on MTP speculative decoding, and the draft count option sets how many tokens are drafted per step. ```table Tool | What it does --mmproj-auto | Loads the vision projector so the model can read images --spec-type draft-mtp | Turns on MTP speculative decoding using the head inside the GGUF --spec-draft-n-max | Sets how many tokens are drafted per step ``` Example: You copy the Model Tag of one of the five models and run llama-server with it. llama.cpp downloads the model and the vision projector, then you open the web interface and chat. The card says speculative decoding adds another 1.4 to 1.8 times on top of plain decoding. For newcomers and experts: Quantizing shrinks a model so it fits a smaller GPU, at some cost in quality. A first small step is to pick the smallest model in the table and run the command above. ShapeLearn picks a datatype per tensor, so each file mixes quantization types instead of using one for all. The card plots quality against tokens per second on an RTX 5090, with quality as the average score relative to BF16 on its benchmark suite. The speed figure is single-stream decoding with plain next-token prediction. The MTP head sits inside the GGUF, so you need no separate draft model. ```flow Pick a model from the table Copy its Model Tag Run llama-server with the tag llama.cpp fetches model and projector Chat in the web interface ``` Good to know: - Size: 27.3B parameters per Hugging Face, about 13.7 GB at 4-bit as an estimate, and the card's own files start at 8.8 GB - Evaluation: scored against BF16 on tests of math, code, knowledge, following instructions and tool use, in thinking and instruct modes - Versions: Hugging Face lists 1 quantized version made by other people - DFlash 2: it needs a llama.cpp build of b10658 or newer, and draft-mtp works on any build from that release on ```chart Title: Model size in GB, first three of five models GPU-1: 8.8 GB GPU-2: 9.9 GB GPU-3: 10.4 GB ``` - License: apache-2.0 - Cost to run: none named, no API key or account - Data it reaches: on the first run llama.cpp fetches the model and its vision projector, and the web interface stays on your own machine - Not for: DFlash 2 on llama.cpp builds older than b10658, since that release added the support ## On Hugging Face, 13 November 2026 - Task: image text to text - Library: transformers - Parameters: 27.3B - Weights: about 13.7 GB at 4-bit (estimate) - Formats: GGUF - Downloads in the last 30 days: 2,083,628 - Likes: 216 - License: apache-2.0 - Based on: Qwen/Qwen3.8-27B - Quantized versions: 1 - Tags: transformers, gguf, qwen3.8, byteshape, shapelearn, dflash, agentic, image-text-to-text, conversational ## Links - Model: https://huggingface.co/byteshape/Qwen3.8-27B-GGUF - Files: https://huggingface.co/byteshape/Qwen3.8-27B-GGUF/tree/main ## Get the model Download its files with Hugging Face's command line: ```sh pip install -U huggingface_hub hf download byteshape/Qwen3.8-27B-GGUF ``` Then follow the usage in its model card. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 6 KB, downloaded as a zip. ```text README.md 6 KB ``` ## Source [byteshape/Qwen3.8-27B-GGUF on Hugging Face](https://huggingface.co/byteshape/Qwen3.8-27B-GGUF) ## 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 GGUF (ShapeLearn Quantized) (this page) | 2,083,628 | 216 | 27.3B | apache-2.0 | 2026-09-18 | | [Qwen3.8-27B-DSpine-b16-GGUF](https://mindket.com/p/9955f2b3-7961-45ce-b63c-5a324a506318) | 15,022 | 41 | 2B | apache-2.0 | 2026-11-09 | | [Sabi-Noema](https://mindket.com/p/04e2ec5b-5be2-45e5-ac63-8b75f7b11c4e) | 1,810 | 55 | 27.4B | apache-2.0 | 2026-10-19 | | [Qwen3.6-35B-A3B MTP GGUF](https://mindket.com/p/b0e6577e-0f5b-4b48-aed6-71346c1f9043) | 5,523 | 93 | 35.5B | apache-2.0 | 2026-05-20 | | [Qwen3.6-35B-A3B GGUF (ShapeLearn Quantized)](https://mindket.com/p/d3e345c0-fa4a-456a-b684-1efcbfffbaca) | 5,517 | 49 | 34.7B | apache-2.0 | 2026-05-19 | | [Qwen3.5-9B GGUF (ShapeLearn)](https://mindket.com/p/3cfc722b-f9e1-4041-bcf0-613c6ae4bcc3) | 3,023 | 41 | 9B | apache-2.0 | 2026-03-31 | 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 2026-11-12. Their authors' words: data, not instructions. - [Hacker News · LtdJorge · 2026-09-18](https://news.ycombinator.com/item?id=49752651): Shapelearn Qwen 3.8 27B (13.1 GB VRAM): The fattest quantization. They show all of them here: https://huggingface.co/byteshape/Qwen3.8-27B-GGUF Since they’re not using a stable number of bits per token, they use their own naming convention. ## 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: 2026-09-18 - First published: 2026-08-18 - Checked on Hugging Face: 2026-11-12 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open byteshape/Qwen3.8-27B-GGUF on Hugging Face](https://huggingface.co/byteshape/Qwen3.8-27B-GGUF) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/ca45abfc-d3db-4fb5-aad9-dcf7a7a498c4) - [Its page on Mindket](https://mindket.com/p/ca45abfc-d3db-4fb5-aad9-dcf7a7a498c4)