{"slug": "qwen3-8-flash-next-how-to-run-locally", "title": "Qwen3.8-Flash-Next: How to Run Locally", "summary": "Qwen released Qwen3.8-Flash-Next, a 125B-parameter open-weight multimodal MoE model built on the Qwen4 architecture with a 262K context window, which outperforms Claude-4.6-Opus (Max) and can run locally on devices with 75GB RAM or unified memory. Unsloth provides GGUFs and a llama.cpp PR for running the model in Unsloth Desktop, with the 1-bit quantized version at 75GB being 79% smaller than BF16 (355GB) while retaining a top-1% accuracy of 80%.", "body_md": "# Qwen3.8-Flash-Next: How to Run Locally\n\nGuide to run Qwen3.8-Flash-Next locally.\n\nQwen3.8-Flash-Next is a new open-weight, **125B parameter** MoE multimodal model from Qwen. Built on the new Qwen4 architecture, it supports a 262K context window and advanced reasoning. The model outperforms Claude-4.6-Opus (Max) and can run locally on devices with **75GB RAM**/unified memory with no GPU VRAM required. To run the model, use our [GGUFs](https://huggingface.co/unsloth/Qwen3.8-Flash-Next-GGUF) with our specific [llama.cpp PR](https://github.com/ggml-org/llama.cpp/pull/27742) for [Unsloth Desktop](/docs/desktop) as well. Thank you Qwen for day zero access.\n\n**1-bit is 75GB** and uses 4-bit for the Ngram / PLE. This is **79% smaller** than BF16 (355GB), and retains a **top-1% accuracy of 80%**.\n\n[Run Qwen3.8-Flash-Next Guide](/docs/models/qwen3.8-next#run-qwen3.8-flash-next-in-unsloth)[Download Unsloth](https://unsloth.ai/download)\n\nQwen3.8-Flash-Next is now available to run locally in [Unsloth Desktop](/docs/models/qwen3.8-next#run-qwen3.8-flash-next-in-unsloth)!\n\n### ⚙️ Usage Guide\n\nWhether you run **Qwen3.8-Flash-Next** on a CPU with system RAM or on a GPU with VRAM may make relatively little difference. Its unique architecture allows inference using RAM or unified memory to achieve performance closer to that of GPU VRAM than is typical for other models. This makes it particularly well suited to Macs, NVIDIA DGX Spark systems, and other devices with large memory capacities.\n\nYou will need at least **75 GB of RAM or unified memory** to run the model. Its smallest 1-bit quantized version is larger than usual because of new Ngram layers or per layer embeddings which is like a lookup table. However, this also means the quantization is less aggressive, allowing the model to retain more of its original accuracy than more heavily quantized models. You can also offload the PLE / Ngram layer to SSD and use mmap which allows less usage of CPU and GPU VRAM.\n\n#### Qwen3.8-Flash-Next Requirements:\n\nThe smallest quant works on 75GB RAM so it's best to have a 96GB RAM/unified memory device.\n**Table: Hardware requirements** (units = total memory: RAM + VRAM, or unified memory)\n\n75 GB\n\n79 GB\n\n90 GB\n\n112 GB\n\n200 GB\n\n270 GB\n\n355 GB\n\n### Recommended Settings\n\nQwen3.8-Flash-Next is a** hybrid thinking** model with different default settings for thinking and non-thinking modes. Extra high is enabled by default so if you want shorter thinking traces, you can [adjust the thinking effort](/docs/models/qwen3.8-next#thinking--preserve-thinking):\n\n`temperature`\n\n1.0\n\n0.7\n\n`top_p`\n\n0.95\n\n0.80\n\n`top_k`\n\n20\n\n20\n\n`min_p`\n\n0.0\n\n0.0\n\n`presence_penalty`\n\n0.0\n\n1.5\n\n`repetition_penalty`\n\n1.0\n\n1.0\n\nContext length = up to\n\n`262,144`\n\nThinking Mode:\n\n`temperature=1.0`\n\n,`top_p=0.95`\n\n,`top_k=20`\n\n,`min_p=0.0`\n\n,`presence_penalty=0.0`\n\n,`repetition_penalty=1.0`\n\nInstruct (or non-thinking) mode:\n\n`temperature=0.7`\n\n,`top_p=0.80`\n\n,`top_k=20`\n\n,`min_p=0.0`\n\n,`presence_penalty=1.5`\n\n,`repetition_penalty=1.0`\n\n### 💡 Thinking + Preserve Thinking\n\nQwen3.8-Flash-Next has **Preserve Thinking** which leaves the thinking trace from the previous conversation. This increases the number of tokens you use, but could increase accuracy in continued conversations. [Unsloth](/docs/models/qwen3.8-next#run-qwen3.8-in-unsloth-desktop) has 'Think' and Preserved Thinking toggles for Qwen3.8 (see right):\n\nQwen3.8-Flash-Next comes with support for `reasoning_effort`\n\n, which can be used to adjust reasoning depth and control cost. These toggles are automatically enabled in Unsloth:\n\n`xhigh`\n\n(default): for complex tasks demanding thorough analysis`medium`\n\n: balancing accuracy and speed`low`\n\n: efficient reasoning optimizing for speed and costnone\n\nTo change[ thinking / reasoning](/docs/models/qwen3.8-next#how-to-enable-or-disable-reasoning-and-thinking) effort in `unsloth run`\n\nor `llama-server`\n\n, use `--chat-template-kwargs '{\"reasoning_effort\":\"medium\"}'`\n\nIf you're on **Windows** Powershell, use: `--chat-template-kwargs \"{\\\"reasoning_effort\\\":\\\"medium\\\"}\"`\n\nChange `medium`\n\nto your desired reasoning level.\n\n### Quantization Analysis\n\nWe ran KLD for Qwen3.8-Flash quants, and show that 80% top-1% accuracy recovery is possible with 79% less disk space usage. The new architecture uses PLE / Ngrams, and these are not quantized that heavily (4-bit minimum) since they have random access pattern, and quantizing them heavily will damage the model.\n\nUD-Q4_K_XL\n\n111.3\n\n0.044715\n\n93.481\n\nUD-IQ4_XS\n\n93.7\n\n0.079162\n\n91.089\n\nUD-Q3_K_XL\n\n90\n\n0.099694\n\n90.387\n\nUD-IQ3_XXS\n\n82\n\n0.156505\n\n87.57\n\nUD-Q2_K_XL\n\n78.9\n\n0.213343\n\n85.163\n\nUD-IQ1_M\n\n74.5\n\n0.302159\n\n82.396\n\nUD-IQ1_S\n\n72.5\n\n0.37514\n\n80.239\n\n## Run Qwen3.8-Flash-Next Guide\n\nYou can soon run Qwen3.8-Flash-Next in Unsloth Desktop and llama.cpp with our specific PR. Feel free to change quantization type.\n\nHugging Face:\n\n[Qwen3.8-Flash-Next-](https://huggingface.co/unsloth/Qwen3.8-Flash-Next-GGUF)**GGUF** ModelScope:\n\n[Qwen3.8-Flash-Next-GGUF](https://www.modelscope.cn/models/unsloth/Qwen3.8-Flash-Next-GGUF)\n\n[Run in Unsloth Desktop](/docs/models/qwen3.8#run-qwen3.8-in-unsloth-desktop)[Run in llama.cpp](/docs/models/qwen3.8#run-qwen3.8-in-llama.cpp)\n\nQwen3.8-Flash-Next is now available to run locally in [Unsloth Desktop](/docs/models/qwen3.8-next#run-qwen3.8-flash-next-in-unsloth)!\n\n### 🦥 Run Qwen3.8-Flash-Next in Unsloth\n\nQwen3.8-Flash-Next now is able to run in [Unsloth Desktop](/docs/models/qwen3.8-next#run-qwen3.8-in-unsloth-desktop), an open-source UI app for local AI. **Unsloth automatically offloads to RAM and detects multiGPU setups**. With Unsloth Desktop, you can run models locally on **MacOS, Windows**, Linux and:\n\nSearch, download,\n\n[run GGUFs](/docs/new/studio#run-models-locally)and safetensor models+**Self-healing** tool calling**web search**(Python, Bash)** Code execution**[Automatic inference](https://unsloth.ai/docs/desktop#feature-deep-dive)parameter tuning (temp, top-p, etc.)Fast CPU + GPU inference via MLX and llama.cpp\n\n[Train LLMs](/docs/new/studio#no-code-training)2x faster with 70% less VRAM\n\n#### Install Unsloth\n\nThe easiest way to get started is by downloading the [Unsloth Desktop app](/docs/desktop). Works on [macOS](/docs/get-started/install/mac), [Windows](/docs/get-started/install/windows-installation), and [Linux](/docs/get-started/install/linux).\n\nOr, if you prefer to install manually:\n\nMacOS, Linux, WSL:\n\nWindows PowerShell:\n\n#### Search and download Qwen3.8-Flash-Next\n\nGo to [Unsloth Chat](/docs/new/studio/chat) or Model hub and search for Qwen3.8-Flash in the search bar and download your desired model and quant.\n\n#### Run Qwen3.8-Flash-Next\n\nInference parameters should be auto-set when using Unsloth, however you can still change it manually. You can also edit the context length, chat template and other settings.\n\nFor more information, you can view our [Unsloth inference guide](/docs/new/studio/chat).\n\nFor example using Unsloth Desktop with the 397GB Qwen3.8 (-91% smaller) allows you to toggle thinking modes, allow inline canvas, web search and code execution and much more.\n\n#### Serve Qwen3.8-Flash-Next with Unsloth API\n\nYou can use `unsloth run`\n\ncommand and serve Qwen3.8 via an API using `llama-server`\n\nruntime flags, including context sizing, GPU layers, threading, sampling, networking, and tool configuration. For more info see our [API docs](/docs/basics/api) or [unsloth start](/docs/integrations/unsloth-start).\n\n#### Unsloth is now ready\n\nYou can also do many other things with Qwen3.8-Flash-Next via Unsloth Desktop like:\n\n**Connect tools:**[Claude Code](/docs/basics/claude-code),[Codex](/docs/basics/codex),[web search](/docs/new/studio/chat#advanced-web-search),[MCP](/docs/basics/mcp)and more**Train models:** Fine-tune text, diffusion,[embedding](/docs/basics/embedding-finetuning), and more\n\n### 🦙 Run Qwen3.8-Flash-Next in llama.cpp\n\nWe need to use our specific llama.cpp PR [here](https://github.com/ggml-org/llama.cpp/pull/27742). You can follow the build instructions below as well. Change `-DGGML_CUDA=ON`\n\nto `-DGGML_CUDA=OFF`\n\nif you don't have a GPU or just want CPU inference. **For Apple Mac / Metal devices**, set `-DGGML_CUDA=OFF`\n\nthen continue as usual - Metal support is on by default.\n\nTo run the model, you can do:\n\nThen to run it:\n\n### 📊 Benchmarks\n\nFor GGUF quantization benchmarks you can see above for our [quantization analysis](/docs/models/qwen3.8-next#quantization-analysis) or [Dynamic V3.0 article](/docs/basics/dynamic-3.0-ggufs).\n\nLast updated\n\nWas this helpful?", "url": "https://wpnews.pro/news/qwen3-8-flash-next-how-to-run-locally", "canonical_source": "https://unsloth.ai/docs/models/qwen3.8-next", "published_at": "2026-08-27 12:24:11+00:00", "updated_at": "2026-08-27 12:49:07.345665+00:00", "lang": "en", "topics": ["large-language-models", "ai-products", "ai-tools", "ai-infrastructure"], "entities": ["Qwen", "Qwen3.8-Flash-Next", "Unsloth", "Unsloth Desktop", "Claude-4.6-Opus", "llama.cpp", "NVIDIA DGX Spark"], "alternates": {"html": "https://wpnews.pro/news/qwen3-8-flash-next-how-to-run-locally", "markdown": "https://wpnews.pro/news/qwen3-8-flash-next-how-to-run-locally.md", "text": "https://wpnews.pro/news/qwen3-8-flash-next-how-to-run-locally.txt", "jsonld": "https://wpnews.pro/news/qwen3-8-flash-next-how-to-run-locally.jsonld"}}