Create a REST API for the Microsoft/BitNet B1.58 model and integrate it with an Open WebUI

Now I'm writing/semi-complaining lol. Normally I use models from Ollama, and I happened to come across this person's X (Twitter). They have a Microsoft model that people say can run on CPU just fine, and if you use something like M2, it'll be even faster.

Microsoft just a 1-bit LLM with 2B parameters that can run on CPUs like Apple M2.

BitNet b1.58 2B4T outperforms fp LLaMA 3.2 1B while using only 0.4GB memory versus 2GB and processes tokens 40% faster.

100% opensource. pic.twitter.com/kTeqTs6PHd

— Shubham Saboo (@Saboo_Shubham_) April 18, 2025

And that model is microsoft/BitNet b1.58 2B4T. After seeing the news in APR-2025, I waited to see if anyone would try to implement it in Ollama. I saw people asking about it too, but there was still no update.

I waited for a long time until June-2025 and still nothing. Oh well, let me find a way to run it myself from the code then. Initially, I set a simple goal: put the model in a container and find something that can create an endpoint to work with Open WebUI (a web interface for chatting like ChatGPT), So This blog is to document the my experience for this experiment.

I you want to read Thai Version (อ่านได้ที่นี่)

Table of Contents

• Getting Ready to Run microsoft/BitNet
  
  • - Linux (I actually tried this in Docker)
  • - Windows: This one has quite a few steps, for those who like challenges

  
  

• Writing Code to Use Model from Hugging Face
  
  • - Command Line
  • - API

  
  

• Reference

Getting Ready to Run microsoft/BitNet

- Linux (I actually tried this in Docker)

I take a Python image and install according to these instructions:
# Use official Python 3.12 imageFROM python:3.12-slim# Install system dependencies for PyTorch and build toolsRUN apt-get update && \ apt-get install -y --no-install-recommends \ build-essential \ cmake \ git \ curl \ ca-certificates \ libopenblas-dev \ libomp-dev \ libssl-dev \ libffi-dev \ wget \ && rm -rf /var/lib/apt/lists/*# (Optional) Set a working directoryWORKDIR /app# Copy your requirements.txt if you have oneCOPY requirements.txt .RUN pip install --upgrade pip && pip install -r requirements.txt
create a requirements.txt file.
fastapi==0.110.2uvicorn[standard]==0.29.0transformers==4.52.4torch==2.7.0numpy==1.26.4accelerate==0.29.0
Run it normally (standard execution)
# Build the imagedocker build -t python-bitNet .# Run the container with port forwarding and mounting your codedocker run -it -p 8888:8888 -v "$PWD":/app python-bitNet /bin/bash
Use DevContainer (I try this method)

- Windows: This one has quite a few steps, for those who like challenges

I say it's challenging because I tried it and got stuck for 2 weeks lol. On Linux, it's done in a flash. For anyone who wants to try, you need to have the following:

• For Visual Studio, you need to install additional C++ components as follows:

• Running PowerShell won't work - you have to run it in Developer Command Prompt for VS 2022 or Developer PowerShell for VS 2022.
  The regular Terminal doesn't set all the variables like Path properly, so you'll encounter errors like

Error C1083: Cannot open include file: 'algorithm': No such file or directory

Even though you try to set vcvarsall.bat x64, it's like hit or miss - sometimes it works, sometimes it doesn't.

Note: vcvarsall.bat is located in "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvarsall.bat"

• Set python lib .tlb in path > if you don't include it, it will crash

fatal error LNK1104: cannot open file 'python312.lib'Add Python Lib to PATH for running C++ libraries used for AI Inference
And after the environment is ready

• Let set Virtual environment

# Set ENVpython3 -m venv bitnet-env# orpython -m venv bitnet-env

• Activate Virtual environment

# Linuxsource bitnet-env/bin/activate# Windows - Powershell.\bitnet-env\Scripts\Activate.ps1# Windows - CMD.\bitnet-env\Scripts\activate.bat

• Install required libraries according to requirements.txt - if you're using the Linux/Docker approach, these dependencies are already included in the Container image.

fastapi==0.110.2uvicorn[standard]==0.29.0transformers==4.52.4torch==2.7.0numpy==1.26.4accelerate==0.29.0pip install --upgrade pip && pip install -r requirements.txtpip install git+github.com/huggingface/transfo…

Writing Code to Use Model from Hugging Face

After resolving all the ENV issues, let's Coding, I mentioned wanting to connect it with OpenWebUI, so I made 2 versions Command Line version / API version

- Command Line

Try writing code using

• Transformers: For loading pre-trained models from Hugging Face
• PyTorch: For model inference from transformers (at this point, the my machine specs that only use inference but not good for Train/Fine Tune) The points where PyTorch is used in several parts, such as
  - bfloat16: Uses less memory than FP16 but with less precision
  - return_tensors="pt": Specifies PyTorch tensor format
  - to(model.device): Enables GPU acceleration if CUDA is available

import torchfrom transformers import AutoModelForCausalLM, AutoTokenizermodel_id = "microsoft/bitnet-b1.58-2B-4T"# Load tokenizer and modeltokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.bfloat16, force_download=True,)# Apply the chat template + Rolemessages = [ {"role": "system", "content": "You are a Senior Programmer."}, {"role": "user", "content": "Can you help me with a coding problem?"},]prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)chat_input = tokenizer(prompt, return_tensors="pt").to(model.device)# Generate responsechat_outputs = model.generate(**chat_input, max_new_tokens=50)response = tokenizer.decode(chat_outputs[0][chat_input['input_ids'].shape[-1]:], skip_special_tokens=True)print("\nAssistant Response:", response)The Command Line version - you'll see that Windows has many limitations
Another version actually adds a loop and keeps asking continuously until you type "Thank you BITNET" - you can see source code here

- API

Note I didn't research what libraries are available that can make our API directly connect to Open WebUI. Initially

I tried to see what connection standards Open WebUI supports first - for the Text Prompt part, it has OpenAI / Ollama sections.

Now I chose to go with OpenAI API because when I tried playing with dotnet semantic kernel before, it has the /v1/chat/completions pattern, so I tried starting from there and tried adding it in WebUI to see what paths it hits in our code.

From what I tested, I found there are 3 API endpoints at minimum that Open WebUI calls to us:

• /v1/chat/completions
• /v1/models
• /health

For /v1/chat/completions, I just kept adding based on what it complained about + asked AI until I completed all 3 APIs like this.
import datetimeimport timeimport uuidfrom fastapi import FastAPI, Requestfrom fastapi.responses import JSONResponsefrom pydantic import BaseModelfrom typing import List, Dict, Optionalimport torchimport uuidfrom datetime import datetimefrom transformers import AutoModelForCausalLM, AutoTokenizerapp = FastAPI()# Load model and tokenizer at startupmodel_id = "microsoft/bitnet-b1.58-2B-4T"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.bfloat16, force_download=True,)device = "cuda" if torch.cuda.is_available() else "cpu"model = model.to(device)class Message(BaseModel): role: str content: strclass ChatRequest(BaseModel): messages: List[Message] max_new_tokens: Optional[int] = 700class Choice(BaseModel): index: int message: Dict[str, str] finish_reason: strclass ChatResponse(BaseModel): id: str object: str created: int model: str choices: List[Choice]@app.post("/v1/chat/completions", response_model=ChatResponse)async def chat_completions(request: ChatRequest): # Prepare prompt using chat template prompt = tokenizer.apply_chat_template( [msg.dict() for msg in request.messages], tokenize=False, add_generation_prompt=True ) chat_input = tokenizer(prompt, return_tensors="pt").to(model.device) chat_outputs = model.generate(**chat_input, max_new_tokens=request.max_new_tokens) response = tokenizer.decode( chat_outputs[0][chat_input['input_ids'].shape[-1]:], skip_special_tokens=True ) # Return response in OpenAI-compatible format # return JSONResponse({ # "id": f"chatcmpl-{uuid.uuid4().hex[:12]}", # "object": "chat.completion", # "created": int(time.time()), # "model": model_id, # "choices": [ # { # "index": 0, # "message": { # "role": "assistant", # "content": response # }, # "finish_reason": "stop" # } # ] # }) return ChatResponse( id=f"chatcmpl-{uuid.uuid4().hex[:12]}", object="chat.completion", created=int(time.time()), model=model_id, choices=[ Choice( index=0, message={"role": "assistant", "content": response}, finish_reason="stop" ) ] )@app.get("/")def root(): """Root endpoint with API info""" return JSONResponse({ "message": "OpenAI-Compatible API for Open WebUI", "version": "1.0.0", "endpoints": { "models": "/v1/models", "chat": "/v1/chat/completions", "health": "/health" } })@app.get("/health")def health_check(): """Health check endpoint""" return JSONResponse({"status": "healthy", "timestamp": datetime.now().isoformat()})@app.get("/v1/models")def list_models(): """List available models""" return JSONResponse({ "data": [ { "id": model_id, "object": "model", "created": datetime.now().isoformat(), "owned_by": "microsoft", "permission": [] } ] })
When using it, I made it into Docker. During build, I was shocked by the size - almost 10 GB.

Tried using it for real and connecting it with Open WebUI - sometimes it gives okay answers, sometimes it hallucinates lol

But what's for sure is the CPU usage shoots up lol

That concludes my rough trial of running the Model, and if I find something better, I'll write another blog post. Feel free to reach out with suggestions. Oh, and don't force it on Windows - mine got squeezed by WSL2. Taking an old notebook and installing Linux to make a Local AI Inference Engine is still faster.

For all the code, I've uploaded it to Git: github.com/pingkunga/python_mi…

Reference

• github.com/microsoft/BitNet
• onedollarvps.com/blogs/how-to-…

#python #EnglishBlog #SLM #microsoftBitNet #LocalAIModel

Create a REST API for the Microsoft/BitNet B1.58 model and integrate it with an Open WebUI

Now I'm writing/semi-complaining lol. Normally I use models from Ollama, and I happened to come across this person's X (Twitter). They have a Microsoft model that people say can run on CPU just fine, and if you use something like M2, it'll be even faster. Microsoft just a 1-bit LLM with 2B parameters that can run on CPUs like Apple M2.

naiwaen.debuggingsoft.com/2025…

naiwaen@DebuggingSoft

2025-07-23 08:42:43

Create a REST API for the Microsoft/BitNet B1.58 model and integrate it with an Open WebUI

Now I'm writing/semi-complaining lol. Normally I use models from Ollama, and I happened to come across this person's X (Twitter). They have a Microsoft model that people say can run on CPU just fine, and if you use something like M2, it'll be even faster.

Microsoft just a 1-bit LLM with 2B parameters that can run on CPUs like Apple M2.

BitNet b1.58 2B4T outperforms fp LLaMA 3.2 1B while using only 0.4GB memory versus 2GB and processes tokens 40% faster.

100% opensource. pic.twitter.com/kTeqTs6PHd

— Shubham Saboo (@Saboo_Shubham_) April 18, 2025

And that model is microsoft/BitNet b1.58 2B4T. After seeing the news in APR-2025, I waited to see if anyone would try to implement it in Ollama. I saw people asking about it too, but there was still no update.

I waited for a long time until June-2025 and still nothing. Oh well, let me find a way to run it myself from the code then. Initially, I set a simple goal: put the model in a container and find something that can create an endpoint to work with Open WebUI (a web interface for chatting like ChatGPT), So This blog is to document the my experience for this experiment.

I you want to read Thai Version (อ่านได้ที่นี่)

Table of Contents

• Getting Ready to Run microsoft/BitNet
  
  • - Linux (I actually tried this in Docker)
  • - Windows: This one has quite a few steps, for those who like challenges

  
  

• Writing Code to Use Model from Hugging Face
  
  • - Command Line
  • - API

  
  

• Reference

Getting Ready to Run microsoft/BitNet

- Linux (I actually tried this in Docker)

I take a Python image and install according to these instructions:
# Use official Python 3.12 imageFROM python:3.12-slim# Install system dependencies for PyTorch and build toolsRUN apt-get update && \ apt-get install -y --no-install-recommends \ build-essential \ cmake \ git \ curl \ ca-certificates \ libopenblas-dev \ libomp-dev \ libssl-dev \ libffi-dev \ wget \ && rm -rf /var/lib/apt/lists/*# (Optional) Set a working directoryWORKDIR /app# Copy your requirements.txt if you have oneCOPY requirements.txt .RUN pip install --upgrade pip && pip install -r requirements.txt
create a requirements.txt file.
fastapi==0.110.2uvicorn[standard]==0.29.0transformers==4.52.4torch==2.7.0numpy==1.26.4accelerate==0.29.0
Run it normally (standard execution)
# Build the imagedocker build -t python-bitNet .# Run the container with port forwarding and mounting your codedocker run -it -p 8888:8888 -v "$PWD":/app python-bitNet /bin/bash
Use DevContainer (I try this method)

- Windows: This one has quite a few steps, for those who like challenges

I say it's challenging because I tried it and got stuck for 2 weeks lol. On Linux, it's done in a flash. For anyone who wants to try, you need to have the following:

• For Visual Studio, you need to install additional C++ components as follows:

• Running PowerShell won't work - you have to run it in Developer Command Prompt for VS 2022 or Developer PowerShell for VS 2022.
  The regular Terminal doesn't set all the variables like Path properly, so you'll encounter errors like

Error C1083: Cannot open include file: 'algorithm': No such file or directory

Even though you try to set vcvarsall.bat x64, it's like hit or miss - sometimes it works, sometimes it doesn't.

Note: vcvarsall.bat is located in "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvarsall.bat"

• Set python lib .tlb in path > if you don't include it, it will crash

fatal error LNK1104: cannot open file 'python312.lib'Add Python Lib to PATH for running C++ libraries used for AI Inference
And after the environment is ready

• Let set Virtual environment

# Set ENVpython3 -m venv bitnet-env# orpython -m venv bitnet-env

• Activate Virtual environment

# Linuxsource bitnet-env/bin/activate# Windows - Powershell.\bitnet-env\Scripts\Activate.ps1# Windows - CMD.\bitnet-env\Scripts\activate.bat

• Install required libraries according to requirements.txt - if you're using the Linux/Docker approach, these dependencies are already included in the Container image.

fastapi==0.110.2uvicorn[standard]==0.29.0transformers==4.52.4torch==2.7.0numpy==1.26.4accelerate==0.29.0pip install --upgrade pip && pip install -r requirements.txtpip install git+github.com/huggingface/transfo…

Writing Code to Use Model from Hugging Face

After resolving all the ENV issues, let's Coding, I mentioned wanting to connect it with OpenWebUI, so I made 2 versions Command Line version / API version

- Command Line

Try writing code using

• Transformers: For loading pre-trained models from Hugging Face
• PyTorch: For model inference from transformers (at this point, the my machine specs that only use inference but not good for Train/Fine Tune) The points where PyTorch is used in several parts, such as
  - bfloat16: Uses less memory than FP16 but with less precision
  - return_tensors="pt": Specifies PyTorch tensor format
  - to(model.device): Enables GPU acceleration if CUDA is available

import torchfrom transformers import AutoModelForCausalLM, AutoTokenizermodel_id = "microsoft/bitnet-b1.58-2B-4T"# Load tokenizer and modeltokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.bfloat16, force_download=True,)# Apply the chat template + Rolemessages = [ {"role": "system", "content": "You are a Senior Programmer."}, {"role": "user", "content": "Can you help me with a coding problem?"},]prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)chat_input = tokenizer(prompt, return_tensors="pt").to(model.device)# Generate responsechat_outputs = model.generate(**chat_input, max_new_tokens=50)response = tokenizer.decode(chat_outputs[0][chat_input['input_ids'].shape[-1]:], skip_special_tokens=True)print("\nAssistant Response:", response)The Command Line version - you'll see that Windows has many limitations
Another version actually adds a loop and keeps asking continuously until you type "Thank you BITNET" - you can see source code here

- API

Note I didn't research what libraries are available that can make our API directly connect to Open WebUI. Initially

I tried to see what connection standards Open WebUI supports first - for the Text Prompt part, it has OpenAI / Ollama sections.

Now I chose to go with OpenAI API because when I tried playing with dotnet semantic kernel before, it has the /v1/chat/completions pattern, so I tried starting from there and tried adding it in WebUI to see what paths it hits in our code.

From what I tested, I found there are 3 API endpoints at minimum that Open WebUI calls to us:

• /v1/chat/completions
• /v1/models
• /health

For /v1/chat/completions, I just kept adding based on what it complained about + asked AI until I completed all 3 APIs like this.
import datetimeimport timeimport uuidfrom fastapi import FastAPI, Requestfrom fastapi.responses import JSONResponsefrom pydantic import BaseModelfrom typing import List, Dict, Optionalimport torchimport uuidfrom datetime import datetimefrom transformers import AutoModelForCausalLM, AutoTokenizerapp = FastAPI()# Load model and tokenizer at startupmodel_id = "microsoft/bitnet-b1.58-2B-4T"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.bfloat16, force_download=True,)device = "cuda" if torch.cuda.is_available() else "cpu"model = model.to(device)class Message(BaseModel): role: str content: strclass ChatRequest(BaseModel): messages: List[Message] max_new_tokens: Optional[int] = 700class Choice(BaseModel): index: int message: Dict[str, str] finish_reason: strclass ChatResponse(BaseModel): id: str object: str created: int model: str choices: List[Choice]@app.post("/v1/chat/completions", response_model=ChatResponse)async def chat_completions(request: ChatRequest): # Prepare prompt using chat template prompt = tokenizer.apply_chat_template( [msg.dict() for msg in request.messages], tokenize=False, add_generation_prompt=True ) chat_input = tokenizer(prompt, return_tensors="pt").to(model.device) chat_outputs = model.generate(**chat_input, max_new_tokens=request.max_new_tokens) response = tokenizer.decode( chat_outputs[0][chat_input['input_ids'].shape[-1]:], skip_special_tokens=True ) # Return response in OpenAI-compatible format # return JSONResponse({ # "id": f"chatcmpl-{uuid.uuid4().hex[:12]}", # "object": "chat.completion", # "created": int(time.time()), # "model": model_id, # "choices": [ # { # "index": 0, # "message": { # "role": "assistant", # "content": response # }, # "finish_reason": "stop" # } # ] # }) return ChatResponse( id=f"chatcmpl-{uuid.uuid4().hex[:12]}", object="chat.completion", created=int(time.time()), model=model_id, choices=[ Choice( index=0, message={"role": "assistant", "content": response}, finish_reason="stop" ) ] )@app.get("/")def root(): """Root endpoint with API info""" return JSONResponse({ "message": "OpenAI-Compatible API for Open WebUI", "version": "1.0.0", "endpoints": { "models": "/v1/models", "chat": "/v1/chat/completions", "health": "/health" } })@app.get("/health")def health_check(): """Health check endpoint""" return JSONResponse({"status": "healthy", "timestamp": datetime.now().isoformat()})@app.get("/v1/models")def list_models(): """List available models""" return JSONResponse({ "data": [ { "id": model_id, "object": "model", "created": datetime.now().isoformat(), "owned_by": "microsoft", "permission": [] } ] })
When using it, I made it into Docker. During build, I was shocked by the size - almost 10 GB.

Tried using it for real and connecting it with Open WebUI - sometimes it gives okay answers, sometimes it hallucinates lol

But what's for sure is the CPU usage shoots up lol

That concludes my rough trial of running the Model, and if I find something better, I'll write another blog post. Feel free to reach out with suggestions. Oh, and don't force it on Windows - mine got squeezed by WSL2. Taking an old notebook and installing Linux to make a Local AI Inference Engine is still faster.

For all the code, I've uploaded it to Git: github.com/pingkunga/python_mi…

Reference

• github.com/microsoft/BitNet
• onedollarvps.com/blogs/how-to-…

#python #EnglishBlog #SLM #microsoftBitNet #LocalAIModel

เมื่อกี้ตอนกลับบ้านสวนกับคอร์กี้ แล้วก็นึกขึ้นมาได้ว่าหมาในญี่ปุ่นที่เจอ น้ำหนมาตรฐานตามไกด์ไลน์แทบทุกตัว ไม่มี compromise ว่าอ้วนกว่านี้อีกหน่อยจะน่ารัก ถ้าคนไทยมาเห็นโกลเด้น ลาบ คอร์กี้ เชลตี้ฯลฯ แถวนี้ คงบ่นว่าน้องผอมน่าสงสาร

ปล. แต่่แมวน้ำหนักเกินนี่เห็นเรื่อยๆ นะ ถถถ

ชวนมา Debian Release Party ที่กรุงเทพฯ!

มาร่วมฉลองการเปิดตัว Debian Release ใหม่ล่าสุดกับพวกเราที่กรุงเทพฯ!

อะไร: Debian Release Party - มา "กิน ดื่ม พูดคุย" แลกเปลี่ยนประสบการณ์และทำความรู้จักกับเพื่อนๆ ในคอมมูนิตี้ Debian

ที่ไหน: Jojo Soba ชั้น 2 (ดูแผนที่ได้ที่: openstreetmap.org/way/12212361…)

เมื่อไหร่: วันเสาร์ที่ 9 สิงหาคม พ.ศ. 2568 เวลา 14:00 น. เป็นต้นไป

อย่าพลาดโอกาสดีๆ ที่จะได้มารวมตัวกับคนคอเดียวกัน มาร่วมเป็นส่วนหนึ่งของการเฉลิมฉลองนี้กันนะ!

a bit fruity ตอนล่าสุดพูดถึงประเด็น lgbtq/trans rights คือมันมีอากิวเม้นนึงเกิดขึ้นในสังคมปัจจุบันว่าการเคลื่อนไหวของ lgbtq+ ในปัจจุบันเรียกร้อง "มาก" ไปไหม พยายามทำลายสังคมมากไปป่าว open.spotify.com/episode/52bYI…

ฟังแล้วลองตัดบริบทออกนี่รู้สึกว่าเป็น ep ที่ถามคำถามหลายอย่างกับตัวเรามาก

1. การเคลื่อนไหวเพื่อสิทธิพลเมืองทำยังไงถึงจะ~ถูกต้อง~ ทำยังไงให้สังคมยอมรับ การสร้างบทสนทนาเปิดกว้างมีประโยชน์จริงไหม เราควรเสียเวลาพูดคุยกับคนที่จิตใจเต็มไปด้วยอคติ (bigotry) มั้ย

2. เกิดอะไรขึ้นเมื่อเราบรรลุเป้าหมายแล้วแต่คนอื่นๆที่ไม่ได้อยู่ในจุดเดียวกับเรายังไปไม่ถึงจุดหมายนั้น

The New York Times Thinks Gays Have Gone Too Far (with Chase Strangio)

A Bit Fruity with Matt Bernstein · Episode

^{The New York Times Thinks Gays Have Gone Too Far (with Chase Strangio) (Spotify)}

สรุปแผน Debian Release Party

เมือง: Bangkok, Thailand
กิจกรรม: กิน ดื่ม พูดคุย
สถานที่: Jojo Cafe
เวลา: 14:00น. 9 สิงหาคม พ.ศ. 2568

แต่ว่ายังสมัคร wiki.debian.org/ReleasePartyTr… ไม่ได้ ถ้าเป็นไปได้และถ้าคุณ @thep เข้าได้อยู่แล้วอยากจะขอรบกวนด้วยครับ

ตัวอย่าง TV Anime "Souzai Saishuka no Isekai Ryokouki" โดย Tatsunoko Production x SynergySP เริ่มออกอากาศภายในเดือน ต.ค. 2025

OP: Prologue by Nornis

- Nobunaga Shimazaki➠Takeru
- Ayasa Ito➠Bee
- Makoto Koichi➠Brolite

#souzaisaishuka

As a Meta employee, I can honestly tell you what we know, and I do not know how we obtain all of it.

* Your full name
* Your full home address
* Your phone number
* Your e-mail
* Your government ID
* Your consumer report history
* The name of every family member
* The name of every friend
* The name of their family / friends
* Your marital status
* If you are faithful to your partner
* Your work history (all of it)
* Your education history (all of it)
* Your travel history (going back years)
* Your birth gender
* Your gender ID
* Your sexuality
* Your sexual preferences
* How often you're having sex
* Your partner's details (all the above)
* Your political ideology
* Your involvement with any group
* If you protest, we know
* If you're unhappy, we know

The amount of information we collect on you is insane. And we do it all for supposedly marketing and yes, we help the government since they have access to all this too.

So when someone says they want to avoid META or GOOGLE - respect.

นีโอบัน พลาสเตอร์ยาบรรเทาปวด 3M
yongchieng.com/chinese-medicin…

#ร้านสมุนไพรจีน #สมุนไพร #ยาจีน #ร้านขายยา #ขายยาจีน #ร้านยา #ร้านสมุนไพร #ขายยา #สมุนไพรจีน #ร้านยาจีน #ซื้อยาจีน

'Gachiakuta' เผยโปสเตอร์แอคชัน - Nijiku Zanka

สามาถรับชมได้ที่ YouTube (Ani-One), TrueID, iQIYI, BiliBili TV, Netflix, Crunchyroll

อนิเมะ 'นักสืบตายแล้ว ซีซัน 2' (Tantei wa mou, Shindeiru. Season 2) กำหนดฉายปี 2026/2569

เผยภาพทีเซอร์
กำกับโดย Kurihara Manabu จากสตูดิโอ ENGI

มังงะของอาจารย์ Otsuji ในชื่อ 'คุณแม่ขา ครอบครัวใหม่ไม่รังแกหนู' (Ibitte Konai Gibo to Gishi) ประกาศดัดแปลงเป็นอนิเมะทีวี

เริ่มฉายในปี 2026/2569
กำกับโดย Inoue Keisuke จากสตูดิโอ NEWON

มังงะวางจำหน่ายในไทยโดย เซนชู

Argon are well known for their high-quality Raspberry Pi cases and accessories. Now they're making a 14-inch laptop powered by a Compute Module 5.

omgubuntu.co.uk/2025/07/argon-…

#RaspberryPi #linux #tech

Argon ONE Up Laptop Runs on a Raspberry Pi CM5

Argon ONE UP is a 14-inch laptop powered by a Raspberry Pi Compute Module 5. It boasts ports, expansion ports, and 40 pin GPIO add-on. Details inside.

^{Joey Sneddon (OMG! Ubuntu!)}

อนิเมะ 'Pokémon Concierge Part 2' เริ่มสตรีมบน Netflix ทั่วโลกวันที่ 4 กันยายนนี้

เผยโปสเตอร์ใหม่
กำกับโดย Ogawa Iku จากสตูดิโอ dwarf

Space Dingus is a spiritual sequel to Death Road to Canada gamingonlinux.com/2025/07/spac…

#SpaceDingus #IndieGames #Gaming #PCGaming

Space Dingus is a spiritual sequel to Death Road to Canada

Death Road to Canada is a lot of fun and so I have great expectations for the next game from Rocketcat Games, which is a spiritual sequel named Space Dingus.

^{Liam Dawe (GamingOnLinux)}

Splitgate 2 is "going back to beta" with the original shutting down as 1047 Games have layoffs gamingonlinux.com/2025/07/spli…

#Splitgate2 #Splitgate #FPS #PCGaming #Gaming

Splitgate 2 is "going back to beta" with the original shutting down as 1047 Games have layoffs

1047 Games have announced that Splitgate 2 is going to be reworked, and they're shutting down the original Splitgate.

^{Liam Dawe (GamingOnLinux)}

เอพริล: พ่อเพื่อนคนนี้แด๊ดดี้พันธุ์แซ่บ
ถ้าตอนนี้คุณคือ... เอพริล มอร์แกน ทายาทสาวของมหาเศรษฐีที่มีรูปร่างเย้ายวนชวนมอง ผิวขาวผ่องราวไข่มุก ปากอิ่มแดงราวกลีบกุหลาบ ดวงตากลมโตเป็นประกายท้าทายทุกสิ่ง และชอบทำในสิ่งที่ใจปรารถนา
คุณจะเลือกใคร...คนไหน?
ถ้าสามหนุ่ม สามวัย จากตระกูลขุนนางเก่า เดออน กอนเดอแวน ที่ทั้งเถื่อน ร้าย เร่าร้อน และร้อนแรง มาวอแวตามจีบคุณ

[1] ตอนที่ 1: เปลวไฟใต้เงามืด
แสงไฟนีออนสีม่วงจากคลับหรูละลายรวมกับเสียงดนตรีเร้าใจที่ดังกระหึ่มจนพื้นสะเทือน เอพริล มอร์แกน สูดลมหายใจเข้าลึก กลิ่นแอลกอฮอล์และน้ำหอมราคาแพงเคล้าคละกันจนเวียนหัว แต่เธอกลับยิ้ม #เตรียมพบกันเที่ยงคืนนี้นะคะ #นิยาย #ebook

Now Live!!! Let's talk about cables!!!!

youtube.com/live/7zmFYHgWFLE

Extreme sports game Descenders Next has released into Early Access gamingonlinux.com/2025/07/extr…

#Descenders #DescendersNext #Gaming #PCGaming

Extreme sports game Descenders Next has released into Early Access

Descenders Next is a fresh extreme sports game from RageSquid / No More Robots and it shadow-dropped on Steam in Early Access.

^{Liam Dawe (GamingOnLinux)}

Skip running multiple commands to update packages on Ubuntu with TopGrade, a CLI tool able to upgrade everything, from apt to snap, vscode plugins to pip – all with just a single command.

omgubuntu.co.uk/2025/07/topgra…

#linux #ubuntu #opensource

This Tool Upgrades Everything on Ubuntu Using One Command

Skip running multiple commands to update packages on Ubuntu with TopGrade, a CLI tool which upgrades everything, from apt to pip, with a single command.

^{Joey Sneddon (OMG! Ubuntu!)}

Hellraiser game announced with Clive Barker's Hellraiser: Revival gamingonlinux.com/2025/07/hell…

#Hellraiser #Gaming #PCGaming #HorrorGames

Hellraiser game announced with Clive Barker's Hellraiser: Revival

The first proper Hellraiser game has just been revealed with Clive Barker's Hellraiser: Revival.

^{Liam Dawe (GamingOnLinux)}

Co-op climbing game PEAK is a truly great time with friends gamingonlinux.com/2025/07/co-o…

#PEAK #IndieGames #Gaming #PCGaming #Linux #LinuxGaming

Co-op climbing game PEAK is a truly great time with friends

PEAK is another in the list of quality co-op chaos games to add to your collection.

^{Liam Dawe (GamingOnLinux)}

AI-assisted tab groups, reduced memory usage on Linux, unit conversions in the address bar AND expanded PiP support – it's Firefox 141.

omgubuntu.co.uk/2025/07/firefo…

#firefox #opensource #foss

Firefox 141 Adds AI Tab Grouping, Reduces Memory Use on Linux

Firefox 141 is out, featuring AI-powered tab groups, lower memory usage on Linux, a new unit converter, and expanded PiP support. See what's new.

^{Joey Sneddon (OMG! Ubuntu!)}

Whimsical exploration-trading adventure Townseek gets a big demo refresh gamingonlinux.com/2025/07/whim…

#IndieGames #Demo #PCGaming #Gaming #Steam

Whimsical exploration-trading adventure Townseek gets a big demo refresh

Townseek is a colourful casual exploration and trading adventure game from developer Whales And Games, with a brand new big upgrade for the demo available.

^{Liam Dawe (GamingOnLinux)}