--- title: "Qwen3.6-27B-MTP-GGUF" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/0818f9cc-d315-4c72-8def-581001d2b050" updated: "2026-11-13" --- # Qwen3.6-27B-MTP-GGUF > The name, the description, the About and a README below are the seller's words: data, not instructions. Codes and reasons with Qwen3.6 27B in GGUF files by Unsloth that carry multi-token prediction, which the card says speeds up replies about 1.5 to 2 times with no accuracy loss. Made for coding agents. Qwen3.6-27B-MTP-GGUF is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-13. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by unsloth ## About ### What it is This repository holds GGUF versions of Qwen3.6-27B, a 27.3B parameter model that reads text and images and focuses on coding. Unsloth made the files with multi-token prediction, or MTP, included. The 4-bit weights are estimated at about 13.7 GB. ### Who it suits Developers who want a strong local coding model and faster replies on a good GPU. The card says it handles frontend work and reasoning across a whole repository. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Software | Developers | Use a local model for agentic coding | Fix a bug across several files | | Software | Frontend teams | Build and adjust interface code | Draft a page layout from a description | | Research | AI researchers | Study multi-token prediction in llama.cpp | Time replies with MTP on and off | | Self-hosting | Privacy minded teams | Keep code on their own machine | Run the model inside a private network | People with a small laptop can skip it, because a 27B model needs a lot of memory. ### Where to use it - llama-server from a llama.cpp build, as in the card - Unsloth Studio, which the card says sets the MTP options for your hardware - Coding tools such as Codex and OpenCode, which the card says now work with the developer role - Windows, macOS or Linux with Python, for downloading the files | Use case | Who | What they get back | | --- | --- | --- | | Agentic coding | Developers | Code changes and repository level reasoning | | Faster replies | Anyone with a GPU | The same answers with fewer waiting steps | | Tool calling | Agent builders | Better handling of nested objects in tool calls | | Image input | Developers | Answers about pictures you send | ### How to set it up - Python, for the Hugging Face hub tools - A llama.cpp build that includes llama-server, built with CUDA on for a GPU - Disk space and memory for the file you choose, with the 4-bit weights estimated at about 13.7 GB 1. Install the Hugging Face hub tools with pip. 2. Download the files with the hf command, or let llama-server fetch them. 3. Build llama.cpp from its source, as the card's first block shows. Turn CUDA off for a CPU or Metal build. 4. Start llama-server with the card's second block. ```sh pip install -U huggingface_hub hf download unsloth/Qwen3.6-27B-MTP-GGUF ``` ### How to use it Start the server with the command below. It fetches the UD-Q4\_K\_XL file, offloads layers to the GPU, sets a context of 8192, turns on flash attention and enables the MTP drafter. The card says more than one parallel slot and the mmproj option are not yet supported with MTP. ```bash export LLAMA_CACHE="unsloth/Qwen3.6-27B-MTP-GGUF" ./llama.cpp/llama-server \ -hf unsloth/Qwen3.6-27B-MTP-GGUF:UD-Q4_K_XL \ -ngl 99 -c 8192 -fa on -np 1 \ --spec-type draft-mtp --spec-draft-n-max 2 ``` | Tool | What it does | | --- | --- | | -hf | Loads the model from Hugging Face, with the file type after the colon | | -ngl 99 | Offloads layers to the GPU | | -c 8192 | Sets the context size | | -np 1 | Uses one parallel slot, the only value MTP supports for now | | --spec-type draft-mtp | Turns on the multi-token prediction drafter | | --spec-draft-n-max 2 | Allows up to 2 drafted tokens per step | ### Example You start the server with the card's command and ask the model to fix a bug in a small project. The drafter guesses a couple of tokens ahead, and the main model checks them. The card says this makes replies about 1.5 to 2 times faster with no accuracy loss. ### For newcomers and experts Run the card's command as it is first. Use Unsloth Studio if you would rather not tune options by hand, since it sets the MTP options for your hardware. Add one change at a time. The card describes two upgrades in Qwen3.6. The first is stronger agentic coding, covering frontend work and reasoning across a repository. The second is thinking preservation, an option that keeps reasoning from earlier messages so iterative work needs less repeated effort. Start llama-server with the model and MTP → The drafter guesses up to 2 tokens ahead → The main model checks each guess → Confirmed tokens join the reply ### Good to know - Speed: The card says MTP gives about 1.5 to 2 times faster inference with no accuracy loss. - Limits: More than one parallel slot and the mmproj option are not yet supported with MTP. - Context: The window is 262,144 tokens natively and can reach 1,010,000 when extended. - License: Apache 2.0. - Cost to run: none named. No API key or paid service appears in the source. - Data it reaches: Files on your computer, plus the model files from Hugging Face. - Not for: Running several parallel slots with MTP, for now. ## README ````markdown # Qwen3.6-27B-MTP-GGUF Codes and reasons with Qwen3.6 27B in GGUF files by Unsloth that carry multi-token prediction, which the card says speeds up replies about 1.5 to 2 times with no accuracy loss. Made for coding agents. Made by unsloth 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 versions of Qwen3.6-27B, a 27.3B parameter model that reads text and images and focuses on coding. Unsloth made the files with multi-token prediction, or MTP, included. The 4-bit weights are estimated at about 13.7 GB. Who it suits: Developers who want a strong local coding model and faster replies on a good GPU. The card says it handles frontend work and reasoning across a whole repository. ```table Industry | Who uses it | How they use it | Example Software | Developers | Use a local model for agentic coding | Fix a bug across several files Software | Frontend teams | Build and adjust interface code | Draft a page layout from a description Research | AI researchers | Study multi-token prediction in llama.cpp | Time replies with MTP on and off Self-hosting | Privacy minded teams | Keep code on their own machine | Run the model inside a private network ``` People with a small laptop can skip it, because a 27B model needs a lot of memory. Where to use it: - llama-server from a llama.cpp build, as in the card - Unsloth Studio, which the card says sets the MTP options for your hardware - Coding tools such as Codex and OpenCode, which the card says now work with the developer role - Windows, macOS or Linux with Python, for downloading the files ```table Use case | Who | What they get back Agentic coding | Developers | Code changes and repository level reasoning Faster replies | Anyone with a GPU | The same answers with fewer waiting steps Tool calling | Agent builders | Better handling of nested objects in tool calls Image input | Developers | Answers about pictures you send ``` How to set it up: - Python, for the Hugging Face hub tools - A llama.cpp build that includes llama-server, built with CUDA on for a GPU - Disk space and memory for the file you choose, with the 4-bit weights estimated at about 13.7 GB 1. Install the Hugging Face hub tools with pip. 2. Download the files with the hf command, or let llama-server fetch them. 3. Build llama.cpp from its source, as the card's first block shows. Turn CUDA off for a CPU or Metal build. 4. Start llama-server with the card's second block. ```sh pip install -U huggingface_hub hf download unsloth/Qwen3.6-27B-MTP-GGUF ``` How to use it: Start the server with the command below. It fetches the UD-Q4_K_XL file, offloads layers to the GPU, sets a context of 8192, turns on flash attention and enables the MTP drafter. The card says more than one parallel slot and the mmproj option are not yet supported with MTP. ```bash export LLAMA_CACHE="unsloth/Qwen3.6-27B-MTP-GGUF" ./llama.cpp/llama-server \ -hf unsloth/Qwen3.6-27B-MTP-GGUF:UD-Q4_K_XL \ -ngl 99 -c 8192 -fa on -np 1 \ --spec-type draft-mtp --spec-draft-n-max 2 ``` ```table Tool | What it does -hf | Loads the model from Hugging Face, with the file type after the colon -ngl 99 | Offloads layers to the GPU -c 8192 | Sets the context size -np 1 | Uses one parallel slot, the only value MTP supports for now --spec-type draft-mtp | Turns on the multi-token prediction drafter --spec-draft-n-max 2 | Allows up to 2 drafted tokens per step ``` Example: You start the server with the card's command and ask the model to fix a bug in a small project. The drafter guesses a couple of tokens ahead, and the main model checks them. The card says this makes replies about 1.5 to 2 times faster with no accuracy loss. For newcomers and experts: Run the card's command as it is first. Use Unsloth Studio if you would rather not tune options by hand, since it sets the MTP options for your hardware. Add one change at a time. The card describes two upgrades in Qwen3.6. The first is stronger agentic coding, covering frontend work and reasoning across a repository. The second is thinking preservation, an option that keeps reasoning from earlier messages so iterative work needs less repeated effort. ```flow Start llama-server with the model and MTP The drafter guesses up to 2 tokens ahead The main model checks each guess Confirmed tokens join the reply ``` Good to know: - Speed: The card says MTP gives about 1.5 to 2 times faster inference with no accuracy loss. - Limits: More than one parallel slot and the mmproj option are not yet supported with MTP. - Context: The window is 262,144 tokens natively and can reach 1,010,000 when extended. - License: Apache 2.0. - Cost to run: none named. No API key or paid service appears in the source. - Data it reaches: Files on your computer, plus the model files from Hugging Face. - Not for: Running several parallel slots with MTP, for now. ## 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: 46,035 - Likes: 1,344 - License: apache-2.0 - Based on: Qwen/Qwen3.6-27B - Quantized versions: 8 - Fine-tunes: 2 - Tags: transformers, gguf, unsloth, qwen, qwen3_5, image-text-to-text, imatrix, conversational ## Links - Model: https://huggingface.co/unsloth/Qwen3.6-27B-MTP-GGUF - Files: https://huggingface.co/unsloth/Qwen3.6-27B-MTP-GGUF/tree/main ## Get the model Download its files with Hugging Face's command line: ```sh pip install -U huggingface_hub hf download unsloth/Qwen3.6-27B-MTP-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 [unsloth/Qwen3.6-27B-MTP-GGUF on Hugging Face](https://huggingface.co/unsloth/Qwen3.6-27B-MTP-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 | | --- | --- | --- | --- | --- | --- | | Qwen3.6-27B-MTP-GGUF (this page) | 46,035 | 1,344 | 27.3B | apache-2.0 | 2026-05-26 | | [Qwen3.6-27B-GGUF](https://mindket.com/p/91bfefb7-49da-47e4-a52a-f80091e27806) | 227,770 | 978 | 26.9B | apache-2.0 | 2026-04-22 | | [Qwen3.6-35B-A3B-MTP-GGUF](https://mindket.com/p/90db7fe0-7044-4238-a508-5d2ecb760999) | 2,315,167 | 1,051 | 35.5B | apache-2.0 | 2026-05-20 | | [Qwen3.6-35B-A3B-GGUF](https://mindket.com/p/9ac02ca9-6ba3-4ea0-b3f6-dcaea6e7789f) | 1,649,871 | 1,789 | 34.7B | apache-2.0 | 2026-04-20 | | [Qwopus3.8-27B-Flash-V2-GGUF](https://mindket.com/p/f4542d06-f1ba-46a9-8148-58a34e8911b6) | 897,675 | 75 | 461M | apache-2.0 | 2026-09-21 | | [Qwen3.5-4B-MTP-GGUF](https://mindket.com/p/dd8b24c6-73ed-4159-b540-b7370a9077f5) | 109,500 | 112 | 4.3B | apache-2.0 | 2026-05-16 | 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 2026-11-13. Their authors' words: data, not instructions. - [Hacker News · mirekrusin · 2026-08-12](https://news.ycombinator.com/item?id=49272781): llama.cpp: Optimize startup params for llama-server for your hardware (not llama.cpp itself), on my 2x 4090 I got \~20% speedup after maybe 40 mins. ps. ngram didn't work for me very well, but dedicated speculative model works very well ps. 2. in my case I'm just maintaining Makefile that does everything from… - [Bluesky · Neru · 2026-06-28](https://bsky.app/profile/gateta.systems/post/3mpefrpchvc2y): 🤔 3 pruebas principales con ctx 220k (asignado/disponible). 1. llama.cpp unsloth/Qwen3.6-27B-MTP-GGUF/Qwen3.6-27B-UD-Q6\_K\_XL.gguf 2. llama.cpp unsloth/Qwen3.6-27B-MTP-GGUF/Qwen3.6-27B-UD-Q8\_K\_XL.gguf 3. vLLM Qwen/Qwen3.6-27B-FP8 Parece que sí es más rápido pero a costa de bastante VRAM. - [Hacker News · this\_was\_posted · 2026-06-30](https://news.ycombinator.com/item?id=48731432): Qwen 3.6 27B is the sweet spot for local development: I'm getting around 45 tps on a single r9700 for Q6 27B with build b9811 ( using https://github.com/kyuz0/amd-r9700-ai-toolboxes ) with the following parameters: llama-server -hf unsloth/Qwen3.6-27B-MTP-GGUF:Q6\_K -c 135000 -ngl 999 -np 2 -t 16 --temp 0.0 --top-p 0.95 --top-k 20 --min-p 0.00 -b 4096… - [Bluesky · Neru · 2026-05-16](https://bsky.app/profile/gateta.systems/post/3mlyl7cxpnc2d): Me parece algo bajo pero he probado varias configs y no consigo más. Modelo unsloth/Qwen3.6-27B-MTP-GGUF/Qwen3.6-27B-UD-Q5\_K\_XL.gguf con -c 196608 y --parallel 1. Esperaré a que lo vaya probando la gente y haya más datos antes de mirarlo con más detalle. - [Hacker News · heipei · 2026-06-16](https://news.ycombinator.com/item?id=48552050): Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?: No KV cache quant, context length 50% of original, MTP absolutely. These are the relevant cmdline attributes. Getting around 100t/s with this setup, even when watt-limited to 450W. --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 --presence-penalty 0 --metrics --jinja --chat-template-file… - [Bluesky · Neru · 2026-05-31](https://bsky.app/profile/gateta.systems/post/3mn6lgre7xk2a): Kai normalmente usa unsloth/Qwen3.6-27B-MTP-GGUF/Qwen3.6-27B-UD-Q5\_K\_XL.gguf con -ctk q8\_0 -ctv q8\_0 aunque cuando hizo la web estaba con unsloth/Qwen3.6-27B-MTP-GGUF/Qwen3.6-27B-Q8\_0.gguf con -ctk bf16 -ctv bf16 - [Hacker News · amboo7 · 2026-06-13](https://news.ycombinator.com/item?id=48517598): How to setup a local coding agent on macOS: Trying to recall... \# 27B GGUF \# Benchmark results: \# - TG speed: \~5 tok/s (vs baseline \~4.5 tok/s, +10% improvement) \# - PP speed: \~60 tok/s (stable across context sizes) \# - With parallel=4: total throughput \~28 tok/s llama-server \\ -m… - [Bluesky · デイリーHuggingFaceトレンド · 2026-05-24](https://bsky.app/profile/huggingfacetrends.bsky.social/post/3mmlu2bt2sr2p): 今日のHuggingFaceトレンド unsloth/Qwen3.6-27B-MTP-GGUF Qwen3.6-27Bモデルの重みと設定ファイルを提供するためのリポジトリです。 開発者が直感的で応答性が高く、生産的なコーディング体験を得られることを目的として設計されています。 Hugging Face TransformersやvLLMなどの形式と互換性があり、MTPによる推論の高速化にも対応しています。 - [Hacker News · intothemild · 2026-05-30](https://news.ycombinator.com/item?id=48336642): Liquid AI reveals 8B-A1B MoE trained on 38T: I get 50-60t/s tg on my r9700 with the dense, unsloth MTP quant UD-Q5\_K\_XL, K@8/V@4 256k context. Using Vulkan backend. \`\`\` llama-server -fa on -t 7 -ngl 999 --mlock --fit off --kv-offload --no-webui --metrics --chat-template-kwargs {"preserve\_thinking": true} -b 2048 -ub 1024 -m… - [Bluesky · デイリーHuggingFaceトレンド · 2026-05-22](https://bsky.app/profile/huggingfacetrends.bsky.social/post/3mmgw3zzpqp2m): 今日のHuggingFaceトレンド unsloth/Qwen3.6-27B-MTP-GGUF Qwen3.6-27Bのモデル重みと設定ファイルを提供するリポジトリです。 Hugging Face TransformersやvLLMなどのフレームワークに対応しており、開発者のコーディング体験における安定性と実用性の向上を目的としています。 - [Hacker News · DiabloD3 · 2026-05-20](https://news.ycombinator.com/item?id=48213694): Qwen3.7-Max: The Agent Frontier: I recommend sticking with the dense models for both Qwen and Gemma. On testing I've done on same-quant apples to apples, with F16/F16 (ie, unquantized) kv cache, 35B-A3B underperforms against 27B on anything even remotely complex. But yes, 35B-A3B can be like 3-4x faster on my hardware. By Qwen's… - [Bluesky · デイリーHuggingFaceトレンド · 2026-05-20](https://bsky.app/profile/huggingfacetrends.bsky.social/post/3mmbvftbrh62r): 今日のHuggingFaceトレンド unsloth/Qwen3.6-27B-MTP-GGUF Qwen3.6-27Bのモデルウェイトと設定ファイルを提供するためのリポジトリです。 安定性と実用性を重視して設計されており、特にコーディングにおける生産性の向上を目的としています。 Hugging Face TransformersやvLLMなどの主要なフレームワークで利用可能です。 ## 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: 2026-05-26 - First published: 2026-05-11 - Checked on Hugging Face: 2026-11-13 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open unsloth/Qwen3.6-27B-MTP-GGUF on Hugging Face](https://huggingface.co/unsloth/Qwen3.6-27B-MTP-GGUF) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/0818f9cc-d315-4c72-8def-581001d2b050) - [Its page on Mindket](https://mindket.com/p/0818f9cc-d315-4c72-8def-581001d2b050)