Build ollama from source on Windows

Install prerequisites

Install prerequisites:

• Go
• C/C++ Compiler e.g. Clang on macOS, TDM-GCC (Windows amd64) or llvm-mingw (Windows arm64), GCC/Clang on Linux.

Then build and run Ollama from the root directory of the repository:

1

go run . serve

configure and build the project and run Ollama

Install prerequisites:

• CMake
• Visual Studio 2022 including the Native Desktop Workload
• (Optional) NVIDIA GPU support
  • CUDA SDK

[Notice] Ensure prerequisites are in PATH before running CMake.

[Notice] CUDA is only compatible with Visual Studio CMake generators.

Then, configure and build the project:

1
2

cmake -B build
cmake --build build --config Release

Lastly, run Ollama:

1

go run . serve

Build Ollama.exe

Install source code:

1
2

git clone https://github.com/tanwubin/ollama.git
cd ollama

compile the source to ollama.exe

1
2

go generate ./...
go build .

Pull Models

Install Models:

1
2
3

cd ollama
ollama.exe pull llama3.2
ollama.exe pull deepseek-r1:7b

Use curl to test

1
2
3
4
5

curl http://127.0.0.1:11434/api/generate -d '{
  "model": "llama3.2",
  "prompt": "Why is the sky blue?",
  "stream": false
}'

if you see the response of llama3.2 model, it means works!