💜Qwen3.5 - How to Run Locally Guide

Run the new Qwen3.5 LLMs including Medium: Qwen3.5-35B-A3B, 27B, 122B-A10B, Small: Qwen3.5-0.8B, 2B, 4B, 9B and 397B-A17B on your local device!

Qwen3.5 is Alibaba’s new model family, including Qwen3.5-35B-A3B, 27B, 122B-A10B and 397B-A17B and the new Small series: Qwen3.5-0.8B, 2B, 4B and 9B. The multimodal hybrid reasoning LLMs deliver the strongest performances for their sizes. They support 256K context across 201 languages, have thinking + non-thinking, and excel in agentic coding, vision, chat, and long-context tasks. The 35B and 27B models work on a 22GB Mac / RAM device. See all GGUFs here.

Mar 5 Update: Redownload Qwen3.5-35B, 27B and 122B. Redownload 397B once it's updated today.

All GGUFs now updated with an improved quantization algorithm.
All use our new imatrix data. See some improvements in chat, coding, long context, and tool-calling use-cases.
Tool-calling improved following our chat template fixes. Fix is universal and applies to any Qwen3.5 format and any uploader.
Check new GGUF benchmarks for Unsloth performance results + our MXFP4 investigation.
We're retiring MXFP4 layers from 3 Qwen3.5 GGUFs: Q2_K_XL, Q3_K_XL and Q4_K_XL.

All uploads use Unsloth Dynamic 2.0 for SOTA quantization performance - so 4-bit has important layers upcasted to 8 or 16-bit. Thank you Qwen for providing Unsloth with day zero access. You can also fine-tune Qwen3.5 with Unsloth.

To enable or disable thinking see How to enable or disable reasoning & thinking.Qwen3.5 Small models disables by default. Also see LM Studio guide to enable Think toggle.

35B-A3B 27B 122B-A10B 397B-A17B Fine-tune Qwen3.5 0.8B • 2B • 4B • 9B

⚙️ Usage Guide

Table: Inference hardware requirements (units = total memory: RAM + VRAM, or unified memory)

Qwen3.5

3-bit

4-bit

6-bit

8-bit

BF16

0.8B + 2B

3 GB

3.5 GB

5 GB

7.5 GB

9 GB

4.5 GB

5.5 GB

7 GB

10 GB

14 GB

5.5 GB

6.5 GB

9 GB

13 GB

19 GB

27B

14 GB

17 GB

24 GB

30 GB

54 GB

35B-A3B

17 GB

22 GB

30 GB

38 GB

70 GB

122B-A10B

60 GB

70 GB

106 GB

132 GB

245 GB

397B-A17B

180 GB

214 GB

340 GB

512 GB

810 GB

For best performance, make sure your total available memory (VRAM + system RAM) exceeds the size of the quantized model file you’re downloading. If it doesn’t, llama.cpp can still run via SSD/HDD offloading, but inference will be slower.

Between 27B and 35B-A3B, use 27B if you want slightly more accurate results and can't fit in your device. Go for 35B-A3B if you want much faster inference.

Recommended Settings

Maximum context window: 262,144 (can be extended to 1M via YaRN)
presence_penalty = 0.0 to 2.0 default this is off, but to reduce repetitions, you can use this, however using a higher value may result in slight decrease in performance
Adequate Output Length: 32,768 tokens for most queries

If you're getting gibberish, your context length might be set too low. Or try using --cache-type-k bf16 --cache-type-v bf16 which might help.

As Qwen3.5 is hybrid reasoning, thinking and non-thinking mode have different settings:

Thinking mode:

General tasks

Precise coding tasks (e.g. WebDev)

temperature = 1.0

temperature = 0.6

top_p = 0.95

top_k = 20

min_p = 0.0

presence_penalty = 1.5

presence_penalty = 0.0

repeat penalty = disabled or 1.0

Thinking mode for general tasks:

temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0

Thinking mode for precise coding tasks:

temperature=0.6, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=0.0, repetition_penalty=1.0

Instruct (non-thinking) mode settings:

General tasks

Reasoning tasks

temperature = 0.7

temperature = 1.0

top_p = 0.8

top_p = 0.95

top_k = 20

min_p = 0.0

presence_penalty = 1.5

repeat penalty = disabled or 1.0

To disable thinking / reasoning, use --chat-template-kwargs '{"enable_thinking":false}'

If you're on Windows Powershell, use: --chat-template-kwargs "{\"enable_thinking\":false}"

Use 'true' and 'false' interchangeably.

For Qwen3.5 0.8B, 2B, 4B and 9B, reasoning is disabled by default. To enable it, use: --chat-template-kwargs '{"enable_thinking":true}'

Instruct (non-thinking) for general tasks:

temperature=0.7, top_p=0.8, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0

Instruct (non-thinking) for reasoning tasks:

temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0

Qwen3.5 Inference Tutorials:

Because Qwen3.5 comes in many different sizes, we'll be using Dynamic 4-bit MXFP4_MOE GGUF variants for all inference workloads. Click below to navigate to designated model instructions:

Qwen3.5-35B-A3B 27B 122B-A10B 397B-A17B Small (0.8B • 2B • 4B • 9B)LM Studio

Unsloth Dynamic GGUF uploads:

Qwen3.5-35B-A3B

Qwen3.5-27B

Qwen3.5-122B-A10B

Qwen3.5-397B-A17B

presence_penalty = 0.0 to 2.0 default this is off, but to reduce repetitions, you can use this, however using a higher value may result in slight decrease in performance.

Currently no Qwen3.5 GGUF works in Ollama due to separate mmproj vision files. Use llama.cpp compatible backends.

Qwen3.5-35B-A3B

For this guide we will be utilizing Dynamic 4-bit which works great on a 24GB RAM / Mac device for fast inference. Because the model is only around 72GB at full F16 precision, we won't need to worry much about performance. GGUF: Qwen3.5-35B-A3B-GGUF

🦙 Llama.cpp Guides

For these tutorials, we will using llama.cpp for fast local inference, especially if you have a CPU.

Obtain the latest llama.cpp on GitHub here. You can follow the build instructions below as well. Change -DGGML_CUDA=ON to -DGGML_CUDA=OFF if you don't have a GPU or just want CPU inference.

apt-get update
apt-get install pciutils build-essential cmake curl libcurl4-openssl-dev -y
git clone https://github.com/ggml-org/llama.cpp
cmake llama.cpp -B llama.cpp/build \
    -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --clean-first --target llama-cli llama-mtmd-cli llama-server llama-gguf-split
cp llama.cpp/build/bin/llama-* llama.cpp

If you want to use llama.cpp directly to load models, you can do the below: (:Q4_K_M) is the quantization type. You can also download via Hugging Face (point 3). This is similar to ollama run . Use export LLAMA_CACHE="folder" to force llama.cpp to save to a specific location. The model has a maximum of 256K context length.

Follow one of the specific commands below, according to your use-case:

Thinking mode:

Precise coding tasks (e.g. WebDev):

export LLAMA_CACHE="unsloth/Qwen3.5-35B-A3B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-35B-A3B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 0.6 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00

General tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-35B-A3B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-35B-A3B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00

Non-thinking mode:

General tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-35B-A3B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-35B-A3B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 0.7 \
    --top-p 0.8 \
    --top-k 20 \
    --min-p 0.00 \
    --chat-template-kwargs '{"enable_thinking":false}'

Reasoning tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-35B-A3B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-35B-A3B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00 \
    --chat-template-kwargs '{"enable_thinking":false}'

Download the model via (after installing pip install huggingface_hub hf_transfer ). You can choose Q4_K_M or other quantized versions like UD-Q4_K_XL . We recommend using at least 2-bit dynamic quant UD-Q2_K_XL to balance size and accuracy. If downloads get stuck, see: Hugging Face Hub, XET debugging

hf download unsloth/Qwen3.5-35B-A3B-GGUF \
    --local-dir unsloth/Qwen3.5-35B-A3B-GGUF \
    --include "*UD-Q4_K_XL*" # Use "*UD-Q2_K_XL*" for Dynamic 2bit

Then run the model in conversation mode:

./llama.cpp/llama-cli \
    --model unsloth/Qwen3.5-35B-A3B-GGUF/Qwen3.5-35B-A3B-UD-Q4_K_XL.gguf \
    --mmproj unsloth/Qwen3.5-35B-A3B-GGUF/mmproj-F16.gguf \
    --seed 3407 \
    --temp 1.0 \
    --top-p 0.95 \
    --min-p 0.01 \
    --top-k 40

Qwen3.5 Small (0.8B • 2B • 4B • 9B)

For Qwen3.5 0.8B, 2B, 4B and 9B, reasoning is disabled by default. To enable it, use: --chat-template-kwargs '{"enable_thinking":true}'

On Windows use: --chat-template-kwargs "{\"enable_thinking\":true}"

For the Qwen3.5 Small series, because they're so small, all you need to do is change the model name in the scripts to desired variant. For this specific guide we'll be using the 9B parameter variant. To run them all in near full precision, you'll just need 12GB of RAM / VRAM / unified memory device. GGUFs:

Obtain the latest llama.cpp on GitHub here. You can follow the build instructions below as well. Change -DGGML_CUDA=ON to -DGGML_CUDA=OFF if you don't have a GPU or just want CPU inference.

apt-get update
apt-get install pciutils build-essential cmake curl libcurl4-openssl-dev -y
git clone https://github.com/ggml-org/llama.cpp
cmake llama.cpp -B llama.cpp/build \
    -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --clean-first --target llama-cli llama-mtmd-cli llama-server llama-gguf-split
cp llama.cpp/build/bin/llama-* llama.cpp

If you want to use llama.cpp directly to load models, you can do the below: (:Q4_K_XL) is the quantization type. You can also download via Hugging Face (point 3). This is similar to ollama run . Use export LLAMA_CACHE="folder" to force llama.cpp to save to a specific location. The model has a maximum of 256K context length.

Follow one of the specific commands below, according to your use-case:

To use another variant other than 9B, you can change the '9B' to: 0.8B, 2B or 4B etc.

Thinking mode (disabled by default)

Qwen3.5 Small models disable thinking by default. Use llama-server to enable it.

export LLAMA_CACHE="unsloth/Qwen3.5-9B-GGUF"
./llama.cpp/llama-server \
    -hf unsloth/Qwen3.5-9B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 0.6 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00 \
    --alias "unsloth/Qwen3.5-9B-GGUF" \
    --port 8001 \
    --chat-template-kwargs '{"enable_thinking":true}'

General tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-9B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-9B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00 \
    --alias "unsloth/Qwen3.5-9B-GGUF" \
    --port 8001 \
    --chat-template-kwargs '{"enable_thinking":true}'

To use another variant other than 9B, you can change the '9B' to: 0.8B, 2B or 4B etc.

Non-thinking mode is already on by default

General tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-9B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-9B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 0.7 \
    --top-p 0.8 \
    --top-k 20 \
    --min-p 0.00

Reasoning tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-9B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-9B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00

hf download unsloth/Qwen3.5-9B-GGUF \
    --local-dir unsloth/Qwen3.5-9B-GGUF \
    --include "*UD-Q4_K_XL*" # Use "*UD-Q2_K_XL*" for Dynamic 2bit

Then run the model in conversation mode:

./llama.cpp/llama-cli \
    --model unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-UD-Q4_K_XL.gguf \
    --mmproj unsloth/Qwen3.5-9B-GGUF/mmproj-F16.gguf \
    --seed 3407 \
    --temp 1.0 \
    --top-p 0.95 \
    --min-p 0.01 \
    --top-k 40

Qwen3.5-27B

For this guide we will be utilizing Dynamic 4-bit which works great on a 18GB RAM / Mac device for fast inference. GGUF: Qwen3.5-27B-GGUF

Obtain the latest llama.cpp on GitHub here. You can follow the build instructions below as well. Change -DGGML_CUDA=ON to -DGGML_CUDA=OFF if you don't have a GPU or just want CPU inference.

apt-get update
apt-get install pciutils build-essential cmake curl libcurl4-openssl-dev -y
git clone https://github.com/ggml-org/llama.cpp
cmake llama.cpp -B llama.cpp/build \
    -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --clean-first --target llama-cli llama-mtmd-cli llama-server llama-gguf-split
cp llama.cpp/build/bin/llama-* llama.cpp

Follow one of the specific commands below, according to your use-case:

Thinking mode:

Precise coding tasks (e.g. WebDev):

export LLAMA_CACHE="unsloth/Qwen3.5-27B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-27B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 0.6 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00

General tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-27B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-27B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00

Non-thinking mode:

General tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-27B-GGUF"
./llama.cpp/llama-server \
    -hf unsloth/Qwen3.5-27B-GGUF:Q4_K_M \
    --ctx-size 16384 \
    --temp 0.7 \
    --top-p 0.8 \
    --top-k 20 \
    --min-p 0.00 \
    --chat-template-kwargs '{"enable_thinking":false}'

Reasoning tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-27B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-27B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00 \
    --chat-template-kwargs '{"enable_thinking":false}'

Download the model via (after installing pip install huggingface_hub hf_transfer ). You can choose MXFP4_MOE or other quantized versions like UD-Q4_K_XL . We recommend using at least 2-bit dynamic quant UD-Q2_K_XL to balance size and accuracy. If downloads get stuck, see: Hugging Face Hub, XET debugging

hf download unsloth/Qwen3.5-27B-GGUF \
    --local-dir unsloth/Qwen3.5-27B-GGUF \
    --include "*UD-Q4_K_XL*" # Use "*UD-Q2_K_XL*" for Dynamic 2bit

Then run the model in conversation mode:

./llama.cpp/llama-cli \
    --model unsloth/Qwen3.5-27B-GGUF/Qwen3.5-27B-UD-Q4_K_XL.gguf \
    --mmproj unsloth/Qwen3.5-27B-GGUF/mmproj-F16.gguf \
    --seed 3407 \
    --temp 1.0 \
    --top-p 0.95 \
    --min-p 0.01 \
    --top-k 40

Qwen3.5-122B-A10B

For this guide we will be utilizing Dynamic 4-bit which works great on a 70GB RAM / Mac device for fast inference. GGUF: Qwen3.5-122B-A10B-GGUF

Obtain the latest llama.cpp on GitHub here. You can follow the build instructions below as well. Change -DGGML_CUDA=ON to -DGGML_CUDA=OFF if you don't have a GPU or just want CPU inference.

apt-get update
apt-get install pciutils build-essential cmake curl libcurl4-openssl-dev -y
git clone https://github.com/ggml-org/llama.cpp
cmake llama.cpp -B llama.cpp/build \
    -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --clean-first --target llama-cli llama-mtmd-cli llama-server llama-gguf-split
cp llama.cpp/build/bin/llama-* llama.cpp

Follow one of the specific commands below, according to your use-case:

Thinking mode:

Precise coding tasks (e.g. WebDev):

export LLAMA_CACHE="unsloth/Qwen3.5-122B-A10B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-122B-A10B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 0.6 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00

General tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-122B-A10B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-122B-A10B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00

Non-thinking mode:

General tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-122B-A10B-GGUF"
./llama.cpp/llama-server \
    -hf unsloth/Qwen3.5-122B-A10B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 0.7 \
    --top-p 0.8 \
    --top-k 20 \
    --min-p 0.00 \
    --chat-template-kwargs '{"enable_thinking":false}'

Reasoning tasks:

export LLAMA_CACHE="unsloth/Qwen3.5-122B-A10B-GGUF"
./llama.cpp/llama-server \
    -hf unsloth/Qwen3.5-122B-A10B-GGUF:UD-Q4_K_XL \
    --ctx-size 16384 \
    --temp 1.0 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00 \
    --chat-template-kwargs '{"enable_thinking":false}'

Download the model via (after installing pip install huggingface_hub hf_transfer ). You can choose MXFP4_MOE (dynamic 4bit) or other quantized versions like UD-Q4_K_XL . We recommend using at least 2-bit dynamic quant UD-Q2_K_XL to balance size and accuracy. If downloads get stuck, see: Hugging Face Hub, XET debugging

hf download unsloth/Qwen3.5-122B-A10B-GGUF \
    --local-dir unsloth/Qwen3.5-122B-A10B-GGUF \
    --include "*UD-Q4_K_XL*" # Use "*UD-Q2_K_XL*" for Dynamic 2bit

Then run the model in conversation mode:

./llama.cpp/llama-cli \
    --model unsloth/Qwen3.5-122B-A10B-GGUF/UD-Q4_K_XL/Qwen3.5-122B-A10B-UD-Q4_K_XL-00001-of-00003.gguf \
    --mmproj unsloth/Qwen3.5-122B-A10B-GGUF/mmproj-F16.gguf \
    --ctx-size 16384 \
    --temp 0.6 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00
    --seed 3407

Qwen3.5-397B-A17B

Qwen3.5-397B-A17B is in the same performance tier as Gemini 3 Pro, Claude Opus 4.5, and GPT-5.2. The full 397B checkpoint is ~807GB on disk, but via Unsloth's 397B GGUFs you can run:

3-bit: fits on 192GB RAM systems (e.g., a 192GB Mac)
4-bit (MXFP4): fits on 256GB RAM. Unsloth 4-bit dynamic UD-Q4_K_XL is ~214GB on disk - loads directly on a 256GB M3 Ultra
Runs on a single 24GB GPU + 256GB system RAM via MoE offloading, reaching 25+ tokens/s
8-bit needs ~512GB RAM/VRAM

See 397B quantization benchmarks on how Unsloth GGUFs perform.

Obtain the latest llama.cpp on GitHub here. You can follow the build instructions below as well. Change -DGGML_CUDA=ON to -DGGML_CUDA=OFF if you don't have a GPU or just want CPU inference.

apt-get update
apt-get install pciutils build-essential cmake curl libcurl4-openssl-dev -y
git clone https://github.com/ggml-org/llama.cpp
cmake llama.cpp -B llama.cpp/build \
    -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --clean-first --target llama-cli llama-mtmd-cli llama-server llama-gguf-split
cp llama.cpp/build/bin/llama-* llama.cpp

Follow this for thinking mode:

export LLAMA_CACHE="unsloth/Qwen3.5-397B-A17B-GGUF"
./llama.cpp/llama-cli \
    -hf unsloth/Qwen3.5-397B-A17B-GGUF:Q4_K_M \
    --ctx-size 16384 \
    --temp 0.6 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00

Follow this for non-thinking mode:

export LLAMA_CACHE="unsloth/Qwen3.5-397B-A17B-GGUF"
./llama.cpp/llama-server \
    -hf unsloth/Qwen3.5-397B-A17B-GGUF:Q4_K_M \
    --ctx-size 16384 \
    --temp 0.7 \
    --top-p 0.8 \
    --top-k 20 \
    --min-p 0.00 \
    --chat-template-kwargs '{"enable_thinking":false}'

hf download unsloth/Qwen3.5-397B-A17B-GGUF \
    --local-dir unsloth/Qwen3.5-397B-A17B-GGUF \
    --include "*Q4_K_M*" # Use "*UD-Q2_K_XL*" for Dynamic 2bit

You can edit --threads 32 for the number of CPU threads, --ctx-size 16384 for context length, --n-gpu-layers 2 for GPU offloading on how many layers. Try adjusting it if your GPU goes out of memory. Also remove it if you have CPU only inference.

./llama.cpp/llama-cli \
    --model unsloth/Qwen3.5-397B-A17B-GGUF/Q4_K_M/Qwen3.5-397B-A17B-Q4_K_M-00001-of-00006.gguf \
    --mmproj unsloth/Qwen3.5-397B-A17B-GGUF/mmproj-F16.gguf \
    --ctx-size 16384 \
    --temp 0.6 \
    --top-p 0.95 \
    --top-k 20 \
    --min-p 0.00
    --seed 3407

👾 LM Studio Guide

For this guide, we'll be using LM Studio, a unified UI interface for running LLMs. The '💡Thinking' and 'Non-thinking' toggle may not appear by default so we'll need some extra steps to get it working.

Download LM Studio for your device. Then open Model Search, search for 'unsloth/qwen3.5', and download the GGUF (quant) that you desire.

Thinking Toggle instructions: After downloading, Open your Terminal / PowerShell and try: lms --help. Then if LM Studio appears normally with many commands, run:

lms get unsloth/qwen3.5-4b

This will get a yaml file which enables your GGUF to have the '💡Thinking' and 'Non-thinking' toggle appear. You can change 4b to the desired quant you'd like to have.

Otherwise, you can go to our LM Studio page and download the specific yaml file.

Restart LM Studio, then load your downloaded model (with the specific thinking toggle you downloaded). You should now see the Thinking toggle enabled. Don't forget to set the correct parameters.

🦙 Llama-server serving & OpenAI's completion library

To deploy Qwen3.5-397B-A17B for production, we use llama-server In a new terminal say via tmux, deploy the model via:

./llama.cpp/llama-server \
--model unsloth/Qwen3.5-35B-A3B-GGUF/Qwen3.5-35B-A3B-UD-Q4_K_XL.gguf \
    --mmproj unsloth/Qwen3.5-35B-A3B-GGUF/mmproj-F16.gguf \
    --alias "unsloth/Qwen3.5-35B-A3B" \
    --temp 0.6 \
    --top-p 0.95 \
    --ctx-size 16384 \
    --top-k 20 \
    --min-p 0.00 \
    --port 8001

Then in a new terminal, after doing pip install openai, do:

from openai import OpenAI
import json
openai_client = OpenAI(
    base_url = "http://127.0.0.1:8001/v1",
    api_key = "sk-no-key-required",
)
completion = openai_client.chat.completions.create(
    model = "unsloth/Qwen3.5-397B-A17B",
    messages = [{"role": "user", "content": "Create a Snake game."},],
)
print(completion.choices[0].message.content)

🤔 How to enable or disable reasoning & thinking

For the below commands, you can use 'true' and 'false' interchangeably. To have Think toggle for LM Studio, read our guide.

To disable thinking / reasoning, use within llama-server:

    --chat-template-kwargs '{"enable_thinking":false}'

If you're on Windows or Powershell, use: --chat-template-kwargs "{\"enable_thinking\":false}"

To enable thinking / reasoning, use within llama-server:

    --chat-template-kwargs '{"enable_thinking":true}'

If you're on Windows or Powershell, use: --chat-template-kwargs "{\"enable_thinking\":true}"

For Qwen3.5 0.8B, 2B, 4B and 9B, reasoning is disabled by default. To enable it, use: --chat-template-kwargs '{"enable_thinking":true}'

And on Windows or Powershell: --chat-template-kwargs "{\"enable_thinking\":true}"

As an example for Qwen3.5-9B to enable thinking (default is disabled):

./llama.cpp/llama-server \
    --model unsloth/Qwen3.5-9B-GGUF/Qwen3.5-9B-BF16.gguf \
    --alias "unsloth/Qwen3.5-9B-GGUF" \
    --temp 0.6 \
    --top-p 0.95 \
    --ctx-size 16384 \
    --top-k 20 \
    --min-p 0.00 \
    --port 8001 \
    --chat-template-kwargs '{"enable_thinking":true}'

And then in Python:

from openai import OpenAI
import json
openai_client = OpenAI(
    base_url = "http://127.0.0.1:8001/v1",
    api_key = "sk-no-key-required",
)
completion = openai_client.chat.completions.create(
    model = "unsloth/Qwen3.5-9B-GGUF",
    messages = [{"role": "user", "content": "What is 2+2?"},],
)
print(completion.choices[0].message.content)
print(completion.choices[0].message.reasoning_content)

👨‍💻 OpenAI Codex & Claude Code

To run the model via local coding agentic workloads, you can follow our guide. Just change the model name 'GLM-4.7-Flash' to your desired 'Qwen3.5' variant and ensure you follow the correct Qwen3.5 parameters and usage instructions. Use the llama-server we just set up just then.

Claude Code

OpenAI Codex

After following the instructions for Claude Code for example you will see:

We can then ask say Create a Python game for Chess :

🔨Tool Calling with Qwen3.5

See Tool Calling Guide for more details on how to do tool calling. In a new terminal (if using tmux, use CTRL+B+D), we create some tools like adding 2 numbers, executing Python code, executing Linux functions and much more:

import json, subprocess, random
from typing import Any
def add_number(a: float | str, b: float | str) -> float:
    return float(a) + float(b)
def multiply_number(a: float | str, b: float | str) -> float:
    return float(a) * float(b)
def substract_number(a: float | str, b: float | str) -> float:
    return float(a) - float(b)
def write_a_story() -> str:
    return random.choice([
        "A long time ago in a galaxy far far away...",
        "There were 2 friends who loved sloths and code...",
        "The world was ending because every sloth evolved to have superhuman intelligence...",
        "Unbeknownst to one friend, the other accidentally coded a program to evolve sloths...",
    ])
def terminal(command: str) -> str:
    if "rm" in command or "sudo" in command or "dd" in command or "chmod" in command:
        msg = "Cannot execute 'rm, sudo, dd, chmod' commands since they are dangerous"
        print(msg); return msg
    print(f"Executing terminal command `{command}`")
    try:
        return str(subprocess.run(command, capture_output = True, text = True, shell = True, check = True).stdout)
    except subprocess.CalledProcessError as e:
        return f"Command failed: {e.stderr}"
def python(code: str) -> str:
    data = {}
    exec(code, data)
    del data["__builtins__"]
    return str(data)
MAP_FN = {
    "add_number": add_number,
    "multiply_number": multiply_number,
    "substract_number": substract_number,
    "write_a_story": write_a_story,
    "terminal": terminal,
    "python": python,
}
tools = [
    {
        "type": "function",
        "function": {
            "name": "add_number",
            "description": "Add two numbers.",
            "parameters": {
                "type": "object",
                "properties": {
                    "a": {
                        "type": "string",
                        "description": "The first number.",
                    },
                    "b": {
                        "type": "string",
                        "description": "The second number.",
                    },
                },
                "required": ["a", "b"],
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "multiply_number",
            "description": "Multiply two numbers.",
            "parameters": {
                "type": "object",
                "properties": {
                    "a": {
                        "type": "string",
                        "description": "The first number.",
                    },
                    "b": {
                        "type": "string",
                        "description": "The second number.",
                    },
                },
                "required": ["a", "b"],
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "substract_number",
            "description": "Substract two numbers.",
            "parameters": {
                "type": "object",
                "properties": {
                    "a": {
                        "type": "string",
                        "description": "The first number.",
                    },
                    "b": {
                        "type": "string",
                        "description": "The second number.",
                    },
                },
                "required": ["a", "b"],
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "write_a_story",
            "description": "Writes a random story.",
            "parameters": {
                "type": "object",
                "properties": {},
                "required": [],
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "terminal",
            "description": "Perform operations from the terminal.",
            "parameters": {
                "type": "object",
                "properties": {
                    "command": {
                        "type": "string",
                        "description": "The command you wish to launch, e.g `ls`, `rm`, ...",
                    },
                },
                "required": ["command"],
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "python",
            "description": "Call a Python interpreter with some Python code that will be ran.",
            "parameters": {
                "type": "object",
                "properties": {
                    "code": {
                        "type": "string",
                        "description": "The Python code to run",
                    },
                },
                "required": ["code"],
            },
        },
    },
]

We then use the below functions (copy and paste and execute) which will parse the function calls automatically and call the OpenAI endpoint for any model:

from openai import OpenAI
def unsloth_inference(
    messages,
    temperature = 0.6,
    top_p = 0.95,
    top_k = 20,
    min_p = 0.00,
    repetition_penalty = 1.0,
):
    messages = messages.copy()
    openai_client = OpenAI(
        base_url = "http://127.0.0.1:8001/v1",
        api_key = "sk-no-key-required",
    )
    model_name = next(iter(openai_client.models.list())).id
    print(f"Using model = {model_name}")
    has_tool_calls = True
    original_messages_len = len(messages)
    while has_tool_calls:
        print(f"Current messages = {messages}")
        response = openai_client.chat.completions.create(
            model = model_name,
            messages = messages,
            temperature = temperature,
            top_p = top_p,
            tools = tools if tools else None,
            tool_choice = "auto" if tools else None,
            extra_body = {"top_k": top_k, "min_p": min_p, "repetition_penalty" :repetition_penalty,}
        )
        tool_calls = response.choices[0].message.tool_calls or []
        content = response.choices[0].message.content or ""
        tool_calls_dict = [tc.to_dict() for tc in tool_calls] if tool_calls else tool_calls
        messages.append({"role": "assistant", "tool_calls": tool_calls_dict, "content": content,})
        for tool_call in tool_calls:
            fx, args, _id = tool_call.function.name, tool_call.function.arguments, tool_call.id
            out = MAP_FN[fx](**json.loads(args))
            messages.append({"role": "tool", "tool_call_id": _id, "name": fx, "content": str(out),})
        else:
            has_tool_calls = False
    return messages

After launching Qwen3.5 via llama-server like in Qwen3.5 or see Tool Calling Guide for more details, we then can do some tool calls.

📊 Benchmarks

Unsloth GGUF Benchmarks

We updated Qwen3.5-35B Unsloth Dynamic quants being SOTA on nearly all bits. We did over 150 KL Divergence benchmarks, totally 9TB of GGUFs. We uploaded all research artifacts. We also fixed a tool calling chat template bug (affects all quant uploaders)

All GGUFs now updated with an improved quantization algorithm.
All use our new imatrix data. See some improvements in chat, coding, long context, and tool-calling use-cases.
Qwen3.5-35B-A3B GGUFs are updated to use new fixes (112B, 27B still converting, re-download once they are updated)
99.9% KL Divergence shows SOTA on Pareto Frontier for UD-Q4_K_XL, IQ3_XXS & more.
Retiring MXFP4 from all GGUF quants: Q2_K_XL, Q3_K_XL and Q4_K_XL, except for pure MXFP4_MOE.

READ OUR DETAILED QWEN3.5 ANALYSIS + BENCHMARKS HERE:

Qwen3.5 GGUF Benchmarks

Qwen3.5-397B-A17B Benchmarks

Benjamin Marie (third-party) benchmarked Qwen3.5-397B-A17B using Unsloth GGUFs on a 750-prompt mixed suite (LiveCodeBench v6, MMLU Pro, GPQA, Math500), reporting both overall accuracy and relative error increase (how much more often the quantized model makes mistakes vs. the original).

Key results (accuracy; change vs. original; relative error increase):

Original weights: 81.3%
UD-Q4_K_XL: 80.5% (−0.8 points; +4.3% relative error increase)
UD-Q3_K_XL: 80.7% (−0.6 points; +3.5% relative error increase)

UD-Q4_K_XL and UD-Q3_K_XL stay extremely close to the original, well under a 1-point accuracy drop on this suite, which Ben insinuates that you can sharply reduce memory footprint (~500 GB less) with little to no practical loss on the tested tasks.

How to choose: Q3 scoring slightly higher than Q4 here is completely plausible as normal run-to-run variance at this scale, so treat Q3 and Q4 as effectively similar quality in this benchmark:

Pick Q3 if you want the smallest footprint / best memory savings
Pick Q4 if you want a slightly more conservative option with similar results

All listed quants utilize our dynamic metholodgy. Even UD-IQ2_M uses a the same methodology of dynamic however the conversion process is different to UD-Q2-K-XL where K-XL is usually faster than UD-IQ2_M even though it's bigger, so that is why UD-IQ2_M may perform better than UD-Q2-K-XL.

Official Qwen Benchmarks

Qwen3.5-35B-A3B, 27B and 122B-A10B Benchmarks

Qwen3.5-4B and 9B Benchmarks

Qwen3.5-397B-A17B Benchmarks

PreviousUltra Long Context RL NextQwen3.5 GGUF Benchmarks

Last updated 9 minutes ago

Was this helpful?

hashtag⚙️ Usage Guide

hashtagRecommended Settings

hashtagThinking mode:

hashtagInstruct (non-thinking) mode settings:

hashtagQwen3.5 Inference Tutorials:

hashtagQwen3.5-35B-A3B

hashtag🦙 Llama.cpp Guides

hashtagQwen3.5 Small (0.8B • 2B • 4B • 9B)

hashtagQwen3.5-27B

hashtagQwen3.5-122B-A10B

hashtagQwen3.5-397B-A17B

hashtag👾 LM Studio Guide

hashtag🦙 Llama-server serving & OpenAI's completion library

hashtag🤔 How to enable or disable reasoning & thinking

hashtag👨‍💻 OpenAI Codex & Claude Code

hashtag🔨Tool Calling with Qwen3.5

hashtag📊 Benchmarks

hashtagUnsloth GGUF Benchmarks

hashtagQwen3.5-397B-A17B Benchmarks

hashtagOfficial Qwen Benchmarks

hashtagQwen3.5-35B-A3B, 27B and 122B-A10B Benchmarks

hashtagQwen3.5-4B and 9B Benchmarks

hashtagQwen3.5-397B-A17B Benchmarks

⚙️ Usage Guide

Recommended Settings

Thinking mode:

Instruct (non-thinking) mode settings:

Qwen3.5 Inference Tutorials:

Qwen3.5-35B-A3B

🦙 Llama.cpp Guides

Qwen3.5 Small (0.8B • 2B • 4B • 9B)

Qwen3.5-27B

Qwen3.5-122B-A10B

Qwen3.5-397B-A17B

👾 LM Studio Guide

🦙 Llama-server serving & OpenAI's completion library

🤔 How to enable or disable reasoning & thinking

👨‍💻 OpenAI Codex & Claude Code

🔨Tool Calling with Qwen3.5

📊 Benchmarks

Unsloth GGUF Benchmarks

Qwen3.5-397B-A17B Benchmarks

Official Qwen Benchmarks

Qwen3.5-35B-A3B, 27B and 122B-A10B Benchmarks

Qwen3.5-4B and 9B Benchmarks

Qwen3.5-397B-A17B Benchmarks