---
# System prepended metadata
title: Qwen 3.6 / Gemma 4 / Nemotron 3 Nano Omni x OpenClaw on NVIDIA Spark (GTC Taipei)
---
# Qwen 3.6 / Gemma 4 / Nemotron 3 Nano Omni x OpenClaw on NVIDIA Spark (GTC Taipei)
## Llama.cpp x Openclaw manual setup
You can install llama.cpp locally to run Qwen 3.6, Gemma 4 and Nemotron 3 Nano Omni models. The justification for using llama.cpp is really about controlling the code base with maximizing the chance success when we deal with multiple models and workflow. Ollama is great, but we pray instead of engineering it when things go wrong.
Here are the ways to setup the serving side, and then we can install OpenClaw next once we verified the serving.
```
#based on this https://unsloth.ai/docs/models/gemma-4
sudo apt-get update
sudo apt-get install pciutils build-essential cmake curl libcurl4-openssl-dev -y python3-pip python3-venv
#this is for browser control and audio
sudo apt-get install chromium mpv
git clone https://github.com/ggml-org/llama.cpp
#lock to the version on 4/26 tested by Ray for GTC Taipei
#git -C llama.cpp checkout 5594d132244aeb1bae54dd431e7efbc908f5e3b8
#locked today 5/4/2026 :-- very stable so far
git -C llama.cpp checkout b97ebdc98f6053604a19d861c08d8087601b96e0
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
```
## Qwen 3.6
We will first download Qwen3.6-35B-A3B model. (Currently the top choice for demo focusing on coding capabilities and tool call following)
```
#install HF transfer if you don't have it
python3 -m venv venv
source venv/bin/activate
pip install huggingface_hub hf_transfer
hf download unsloth/Qwen3.6-35B-A3B-GGUF \
--local-dir unsloth/Qwen3.6-35B-A3B-GGUF \
--include "*mmproj-F16*" \
--include "*UD-Q4_K_XL*" # Use "*UD-Q2_K_XL*" for Dynamic 2bit
```
Then, we can serve the model (warning: port 8000 is used across 3 models, so can conflicts with other serving)
```
#use this if not enough memory
#sudo sysctl -w vm.drop_caches=3
./llama.cpp/llama-server \
--model unsloth/Qwen3.6-35B-A3B-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf \
--mmproj unsloth/Qwen3.6-35B-A3B-GGUF/mmproj-F16.gguf \
--alias "unsloth/Qwen3.6-35B-A3B-GGUF" \
--temp 0.6 \
--top-p 0.95 \
--ctx-size 262144 \
--top-k 20 \
--min-p 0.00 \
--port 8000 \
--checkpoint-every-n-tokens 2048 --ctx-checkpoints 64
#The extra checkpoints seems to reduce some lags due to misses.
#Do not enable this for demo, seems to create some corner cases where the
#bot will not work like cron jobs will stuck
#--chat-template-kwargs '{"preserve_thinking":true}'
```
On another terminal, you can test the llama-server making sure it will response.
```
curl http://127.0.0.1:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
"messages": [
{ "role": "user", "content": "Hi" }
]
}'
```
## Gemma 4
Download the model
```
#install HF transfer if you don't have it
python3 -m venv venv
source venv/bin/activate
pip install huggingface_hub hf_transfer
hf download unsloth/gemma-4-26B-A4B-it-GGUF \
--local-dir unsloth/gemma-4-26B-A4B-it-GGUF \
--include "*mmproj-BF16*" \
--include "*UD-Q4_K_XL*" # Use "*UD-Q2_K_XL*" for Dynamic 2bit
```
Serving the model. DO NOT CLOSE THIS TERMINAL! Keep this terminal open the entire time for all demos (and use it as debug screen).
```
#use this if not enough memory
#sudo sysctl -w vm.drop_caches=3
./llama.cpp/llama-server \
--model unsloth/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-UD-Q4_K_XL.gguf \
--mmproj unsloth/gemma-4-26B-A4B-it-GGUF/mmproj-BF16.gguf \
--temp 1.0 \
--top-p 0.95 \
--top-k 64 \
--alias "unsloth/gemma-4-26B-A4B-it-GGUF" \
--port 8000 \
--cache-ram 0 --ctx-checkpoints 1
--chat-template-kwargs '{"reasoning":"on"}'
```
On another terminal, you can test the llama-server making sure it will response.
```
curl http://127.0.0.1:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
"model": "gemma4:26b",
"messages": [
{ "role": "user", "content": "Hi" }
]
}'
```
## Nemotron 3 Nano Omni (Nemotron-3-Nano-30B-A3B-Omni)
Finally, our lovely Nemotron 3 Nano Omni 33B A3B ... Man, the names are getting so long :-1:
I will not recommend using this for openclaw, and only use this for vlm use cases as subagent. Notice that I used port 8001.
```
hf download unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF \
--local-dir unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF \
--include "*mmproj-BF16*" \
--include "*UD-Q4_K_XL*"
```
Serving the model. DO NOT CLOSE THIS TERMINAL! Keep this terminal open the entire time for all demos (and use it as debug screen).
```
#use this if not enough memory
#sudo sysctl -w vm.drop_caches=3
./llama.cpp/llama-server \
--model unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-UD-Q4_K_XL.gguf\
--mmproj unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF/mmproj-BF16.gguf \
--alias "unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF" \
--prio 3 \
--temp 1.0 \
--top-p 1.0 \
--port 8001
```
On another terminal, you can test the llama-server making sure it will response.
```
curl http://127.0.0.1:8001/v1/chat/completions -H "Content-Type: application/json" -d '{
"messages": [
{ "role": "user", "content": "Hi" }
]
}'
```
Now you have 3 models top models ready to serve anytime but running one of the run script above. Cheers!
A few known issues:
1. The cache-ram and ctx checkpoints will burn the ram, make sure you add these, because unsloth didn't document either. (reported on 4/6/2026)
https://www.reddit.com/r/LocalLLaMA/comments/1sdqvbd/comment/oekiv3j/
https://www.reddit.com/r/openclaw/comments/1sb3ezf/ollamagemma4_is_completely_useless_for_openclaw/
Also, we should experiment with the RAM size and checkpoints to see if we can get any performance gain. So far I didn't notice any improvements.
```
--cache-ram 2048 --ctx-checkpoints 2
```
2. Ollama seems to always quit early or never finish the tasks when it's long. Tool calling seems to be broken, and was patched in 0.20.3 but still not good enough for Openclaw demo. Please check before switching to Ollama.
3. The long tool calling is also proven to be challenging, so when we run demo with Gemma 4 be ready to provide additional instructions like ask the bot to continue the work.
I still cannot figure out the best way to avoid timeout on Ollama or when the model is frozen or not. Please let me know.
## Install Openclaw and Configure the Openclaw.json (IMPORTANT)
```
curl -fsSL https://openclaw.ai/install.sh | bash -s -- --no-onboard
#if you have version too new you can install with this
#to avoid demo failure that's been tested -- uninstall and reinstall
#OpenClaw 2026.5.4 (325df3e) — latest stable
openclaw uninstall --all --yes --non-interactive
npx -y openclaw uninstall --all --yes --non-interactive
#npm install -g openclaw@2026.5.7 #current one mostly used
#seems stable.
npm install -g openclaw@2026.5.12
#image processing for vlm
npm install -g sharp
#tts feature
npm install -g node-edge-tts
```
Now the part we need is to use vllm as the model serving and then point to it. You can use the openclaw onboard interface, follow through the vllms setup.
```
openclaw onboard --install-daemon
```
And the most important part is setting up the vLLM as providers, and here is an example screenshot.
Again, you can switch between these 3 models with llama.cpp.
```
unsloth/gemma-4-26B-A4B-it-GGUF
unsloth/NVIDIA-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF
unsloth/Qwen3.6-35B-A3B-GGUF
```
You can use this openclaw.json file as the reference setup:
> WARNING: !! IMPORTANT READS
> 1. Please make sure you match the context size for the model correctly (i.e., 256k max), and increase the maxToken to at least 16k or higher. Otherwise the tool calling or coding examples will break because of early terminations. I have chosen **128k and 16k max token for performance reason** for faster responding demo.
> 2. make sure adding "image to the input field (i.e., the "input":["text", "image"])
And you can replace the models by replacing the names (e.g., from "unsloth/gemma-4-26B-A4B-it-GGUF" to "unsloth/Qwen3.6-35B-A3B-GGUF" vice versa).
e.g., for Qwen 3.6.
```
{
"agents": {
"defaults": {
"timeoutSeconds": 300,
"model": {
"primary": "unsloth/Qwen3.6-35B-A3B-GGUF"
},
"workspace": "/home/nvidia/.openclaw/workspace",
"models": {
"unsloth/Qwen3.6-35B-A3B-GGUF": {}
}
}
},
...
"vllm": {
"baseUrl": "http://127.0.0.1:8000/v1",
"api": "openai-completions",
"apiKey": "VLLM_API_KEY",
"models": [
{
"id": "unsloth/Qwen3.6-35B-A3B-GGUF",
"name": "unsloth/Qwen3.6-35B-A3B-GGUF",
"reasoning": true,
"input": [
"text", "image"
],
"cost": {
"input": 0,
"output": 0,
"cacheRead": 0,
"cacheWrite": 0
},
"contextWindow": 128000,
"maxTokens": 16000
}
]
}
}
...
```
or for Gemma 4:
```
{
"agents": {
"defaults": {
"timeoutSeconds": 300,
"model": {
"primary": "vllm/unsloth/gemma-4-26B-A4B-it-GGUF"
},
"workspace": "/home/nvidia/.openclaw/workspace",
"models": {
"vllm/unsloth/gemma-4-26B-A4B-it-GGUF": {}
}
}
},
...
"vllm": {
"baseUrl": "http://127.0.0.1:8000/v1",
"api": "openai-completions",
"apiKey": "VLLM_API_KEY",
"models": [
{
"id": "unsloth/gemma-4-26B-A4B-it-GGUF",
"name": "unsloth/gemma-4-26B-A4B-it-GGUF",
"reasoning": true,
"input": [
"text", "image"
],
"cost": {
"input": 0,
"output": 0,
"cacheRead": 0,
"cacheWrite": 0
},
"contextWindow": 128000,
"maxTokens": 16000
}
]
}
}
...
```
### Web Search Skill
#### Ollama Web Search
And lastly ensure you enable the tools needed for demo, and in my case I used ollama web search:
The easiest way is have ollama preinstalled, and select the ollama web search as part of the option in here.
```
openclaw configure --section web
ollama login
```
Otherwise, you can try the steps here as last resort.
https://docs.ollama.com/integrations/openclaw
```
curl -fsSL https://ollama.com/install.sh | sh
openclaw plugins install @ollama/openclaw-web-search
ollama login
#restart gateway before trying the demo
openclaw gateway restart
#run to open the browser interface
openclaw dashboard
```
#### Brave Web Search
Alternatively, you can use Brave Search API:
You can get the $5 credit (1000 request per month) for free, and then after that we can pay for subscriptions
https://brave.com/search/api/guides/use-with-openclaw/
https://docs.openclaw.ai/tools/brave-search
And you can use this command to configure the API keys and all.
```
openclaw configure --section web
```
## Really fun prompts
#### 1. Make a pong game and save it on desktop
```
can you write a html app that have a pong game, save it on Desktop with pong
```
#### 2. Get the latest event information and plan for you!
```
ok, now do a full research and find all source code around openclaw, find the painpoints, and save them at the Desktop openclaw-pain folder. (Document in both English and Korean)
```
#### 3. Upgrade the pong game, and make it better!
```
Ok, read the pong file on my Desktop, and refine and make it better 10x! and make it exciting. Save the results back on Desktop and report back to me.
```
This prompt may fail depends if you have approved the sessions (when it ask for spawn): Run the following command to approve them before re-running it.
```
openclaw devices approve
```
or ask to change the theme:
```
Build me a pong with cat inspired theme, and make it fun.
```
#### 4. Email your game to your friends
1. Setup Agentmail
Create an account https://agentmail.to, and then create an inbox.
Get the API Key, save it in .env file under .openclaw directory.
```
npx clawhub@latest install agentmail
#then restart gateway
openclaw gateway restart
#When you are ready, prompt in the chat to setup your email address before asking to send email.
```
Talk to chatbot in Openclaw and give it the instructions and will finalize the setup.
```
You have access to AgentMail — an Email API for Agents.
The llms.txt file is a very good starting point. Read it first, then go from there based on what the user needs.
llms.txt (overview + all doc links): https://docs.agentmail.to/llms.txt
llms-full.txt (complete reference with inline code examples): https://docs.agentmail.to/llms-full.txt
```
#### Mario inspired like games
```
build mario inspired game in HTML, and make sure it got physics
```
And you can keep improving it by asking it to improve it continuously with some features.
```
add lots of details including hands, arms, legs, and more eyes to the character
```
#### Enable VLM!
You can enable VLM by modifying the `openclaw.json`. You need to add "image" as part of the input.
```
"vllm": {
"baseUrl": "http://127.0.0.1:8000/v1",
"api": "openai-completions",
"apiKey": "VLLM_API_KEY",
"models": [
{
"id": "unsloth/gemma-4-26B-A4B-it-GGUF",
"name": "unsloth/gemma-4-26B-A4B-it-GGUF",
"reasoning": false,
"input": [
"text", "image"
],
"cost": {
"input": 0,
"output": 0,
"cacheRead": 0,
"cacheWrite": 0
},
"contextWindow": 128000,
"maxTokens": 8192
}
]
}
```
And you can connect this to webcam with fswebcam with (`apt-get install fswebcam`). Then, you can setup a prompt or instruction to do something like a cron job with webcam easily with the VLM.
### Qwen 3.6 Prompts
There is a huge improvement in game building with Qwen3.6. But it takes a while as often it will try to create an almost perfect game!
```
let's make a game. I want a mario game, so save the work ~/Desktop/Code and code it with html5 and js
```
So when we run these, also be mindful about the copyright, but again a fun try to see how well it can replicate some of the classical gameplays.
#### Draw something in 3D
```
draw a spinning 3D cube with HTML5 and Three.js
```
#### Go Crazy with 3D Graphics or Game
```
Let's write a 3D mario kart game in html5 and three.js and put that on ~/Desktop/Code/mario_kart
```
#### Use Isaac Sim and build quick Physics Demo
You can prompt the engine to read documentations from github (download locally), and use that to drive a simple 3D simluation demo.
#### Meditation application in HTML + Three.js + Audio
```
yea, build something great for mediation, keep the graphics smooth and simple. And add music to background with nice whitenoise.
```
#### Solve CV problems and write highly efficient app
You can prompt the model to solve classical CV tasks like face detection with Webcam.
```
build me a python application that can do face detection on a webcam. hint: use mediapipe
```
Now we can have openclaw to manage a programmable edge device that can be tuned to different CV applications with your own prompt!
#### Lobster Cam!
You can make fun apps with CV skills above, and here you can ask Openclaw to use this skill
First download the skill as a zip file, unzip it.
https://drive.google.com/drive/folders/1oByz-oT3-Rp1hvJQrqGg8blHlTOdOYhj?usp=sharing
Then, tell Openclaw to read and learn this skill
```
Read the lobstercam skill in ~/Download/lobstercam and run it
```
And ask it to remember or install this skill
```
install the lobstercam skill
```
## Add Telegram
Create the Bot: Open Telegram, message @BotFather, and use the /newbot command. Follow instructions to name your bot and receive the API token.
Then on your Spark, go to terminal and type this
```
openclaw configure --section channels
```
Then go to Telegram and type `/start` in your bot.
Then, go back to terminal
```
openclaw pairing list telegram
openclaw pairing approve telegram
```
Done. :+1: You can now text the chatbot, and you will see a new session under telegram.
## Control your web browser and Do anything!
Enable control with debugging on Chromium
```
/snap/bin/chromium --remote-debugging-port=9222 --remote-debugging-address=127.0.0.1
```
Update the openclaw.json file.
```
"browser": {
"cdpUrl": "http://127.0.0.1:9222",
"attachOnly": true,
"profiles": {
"chrome": {
"cdpUrl": "http://127.0.0.1:9222",
"attachOnly": true,
"color": "#4285F4"
}
}
}
```
Lastly restart openclaw
```
openclaw gateway restart
openclaw browser start
```
Then, tell openclaw to try controlling your browser, and will figure it out itself
```
use the built-in browser skill to open browser google and then search for nvidia
```
```
open amazon and find me the engine oil 5w-30 for my BMW
```
## Podcast style, turn content into speech on webchat
Ask openclaw to install the local tts tool. I like node-edge-tts.
```
can you install node-edge-tts
#openclaw should trigger this, if not you can do it manually
#npm install node-edge-tts
```
Once it's all installed and we can play it back with mpv via the TTS. mpv is installed above, if not install it with `apt-get install mpv`
```
Try this:
npx node-edge-tts -t "Hello from NVBot" -f /tmp/test.mp3 && mpv /tmp/test.mp3 &
```
```
ok find today's news and play it back that way
```
Then you should save the skill to make it run quicker next time (minimizing the discovery steps)
```
save this skill
```
Then, openclaw will create that skill file and so next time you ask for podcast it will know what to do quickly.
This is the fully workaround to get TTS working on Webchat interface.
If you have Telegram, you can just use the default TTS built-in skill, and should just work out of the box without using mpv.
```
\tts on
```
This will turn on TTS, and you can see the audio files pop up as media attachment each time you talk to the agent.
https://docs.openclaw.ai/tools/tts
## Alternative Serving to Try Next
We can also simplify the onboarding with Ollama (given the risk I explained above). I have had lots of headaches due to timeout, or tool calling got stopped randomly! So use this if and only if you are only using it for testing or quick validations. There are workarounds on timeout but needed further investigations.
To run on Spark, you can start a quick install of the latest Ollama. Please make sure you run from this script and check the GPU is activated.
```
# Install latest (YOLO/FOMO way -- up to you)
curl -fsSL https://ollama.com/install.sh | sh
# or install with locked version (safe for demo)
#curl -fsSL https://ollama.com/install.sh | OLLAMA_VERSION=0.23.1 sh
#pull all models are great starter for openclaw experiences
#main driver
ollama pull qwen3.6:35b
#vlm and subagents
ollama pull nemotron3:33b
#long reasoning
ollama pull nemotron-3-super
#coding and well-rounded
ollama pull gemma4:26b
```
That script above will provide a simple chatbot interface on terminal and you can see it in action.
```
ollama ps
```
Also, make sure you check the model is running 100% on GPU. If there are any issues. Try repeating steps here and debug:
https://build.nvidia.com/spark/open-webui/sync
When the model is all ready. Now you can run this command to switch the primary/default model. Or simply follow the onboarding here to install openclaw with Ollama together (do not do that if you have openclaw pre-installed).
https://docs.ollama.com/integrations/openclaw
```
openclaw models set ollama/qwen3.6:35b
openclaw gateway restart
```
You should be able to see the new model in the list of available models at Openclaw. And you can revert it back to vllms with this command.
```
openclaw models set vllm/unsloth/Qwen3.6-35B-A3B-GGUF
openclaw gateway restart
```
## Known Issues:
1. Gemma 4:26b still have the issues in tool calling with openclaw, and there are times it will stop early without warning. Please plan your demo carefully when you are using Gemma 4. Will update on this thread next.
2. Qwen 3.6-35b is amazing at coding, but also takes a long while to complete the job (it seems love to make things perfect on one shot). I will recommend starting with simplier prompt with more directions, to avoid the model go all-in with a single prompt for more responsive demo.
3. nemotron3:33b model is not designed for openclaw. It is great for subagent tasks like VLMs and reasoning things in a scene or world.
Ollama is giving bad output for gemma4, and you can see in the coding example with extra space and typos.
### Workarounds and Findings
1. Reset sessions. There are times when Qwen3.6 or Gemme 4 starts to run long or failed to execute on tasks. You can reset session by typing this command in the chat. This will clear out the context, and will help when some demos may have dependencies such as coding from prior conversations.
```
/reset
```
2. Avoid long open ended tasks. The agent does not have limits on its capabilities, thus it can go try do things impossible within some timeframes. For example, 'process 10000 images with VLM'. This will create a long running loop that may eventually failed. We do not have guardrails for this behavior yet (maybe a good use case for Nemoclaw?).
3. VLMs and multiple models. Nemotron-3-Nano-Omni got better throughput for VLM, but not as great for using as the main driver for openclaw. The workaround now is to enable Nemotron-3-Nano-Omni as subagent tasks, and ideally create custom APIs access to the serving. This it our TODO.
### Some benchmarks to consider in token/s (ollama) and tokens to answers
Prompt used: "why is the sky blue?"
This will trigger reasoning by default.
A good starter reference.
`ollama run gemma4:26b --verbose`
```
total duration: 22.044060213s
load duration: 157.290371ms
prompt eval count: 22 token(s)
prompt eval duration: 58.758482ms
prompt eval rate: 374.41 tokens/s
eval count: 1177 token(s)
eval duration: 21.322907413s
eval rate: 55.20 tokens/s
```
`ollama run qwen3.6:35b --verbose`
```
total duration: 25.762780738s
load duration: 128.753395ms
prompt eval count: 16 token(s)
prompt eval duration: 93.703548ms
prompt eval rate: 170.75 tokens/s
eval count: 1328 token(s)
eval duration: 25.176357481s
eval rate: 52.75 tokens/s
```
`ollama run nemotron-3-super --verbose`
```
total duration: 1m1.157069916s
load duration: 93.651335ms
prompt eval count: 23 token(s)
prompt eval duration: 235.60176ms
prompt eval rate: 97.62 tokens/s
eval count: 1028 token(s)
eval duration: 1m0.652425495s
eval rate: 16.95 tokens/s
```
`ollama run nemotron3:33b --verbose`
```
total duration: 6.498845722s
load duration: 94.549186ms
prompt eval count: 23 token(s)
prompt eval duration: 91.564899ms
prompt eval rate: 251.19 tokens/s
eval count: 359 token(s)
eval duration: 6.134583631s
eval rate: 58.52 tokens/s
```
## Clean Up before Cloning Checklist
- [ ] Remove Ollama Private Key (important)
- [ ] Remove .ssh folder private key (important)
- [ ] Openclaw Session History (use /reset)
- [ ] Delete firefox cookies and caches
- [ ] Delete chrome cookies and caches
- [ ] Clean up Desktop any temp files
- [ ] ~/.openclaw/identity/device-auth.json (OpenClaw tokens)
- [ ] ~/.openclaw/devices/paired.json (paired device tokens)
- [ ] ~/.openclaw/exec-approvals.json (exec socket token)
- [ ] ~/.openclaw/openclaw.json (rename token per machine)
- [ ] Remove chromium lock file `rm ~/snap/chromium/common/chromium/Singleton*`