Link: In marimo, go to the user settings and go to the AI tab. From here
you can find and configure Ollama as an AI provider. For local use you
would typically point the base url to `
Reference these resources when working with Windows.
1---2name: windows-23description: Windows4---56# Windows78> Note: Qwen 3 coder is a 30B parameter model requiring at least 24GB of VRAM to run smoothly. More is required for longer context lengths.910Note: Qwen 3 coder is a 30B parameter model requiring at least 24GB of VRAM to run smoothly. More is required for longer context lengths.1112## Available Resources1314### Usage1516### Using with Claude Code1718### Endpoints1920### Models2122### Differences from the Anthropic API2324### Signing in2526### API keys2728### Status codes2930### Error messages3132### Errors that occur while streaming3334### Get started3536### Base URL3738### Example request3940### Libraries4142- **Python**43 - URL: https://github.com/ollama/ollama-python4445- **JavaScript**46 - URL: https://github.com/ollama/ollama-js4748### Versioning4950### Usage5152### Endpoints5354### Models5556### Disabling streaming5758### When to use streaming vs non-streaming5960### Example response6162### Recommended models6364- **embeddinggemma**65 - URL: https://ollama.com/library/embeddinggemma6667- **qwen3-embedding**68 - URL: https://ollama.com/library/qwen3-embedding6970- **all-minilm**71 - URL: https://ollama.com/library/all-minilm7273### Generate embeddings7475### Generate a batch of embeddings7677### Tips7879### Key streaming concepts8081### Handling streamed chunks8283### Generating structured JSON8485### Generating structured JSON with a schema8687### Example: Extract structured data8889### Example: Vision with structured outputs9091### Tips for reliable structured outputs9293### Supported models9495- **Qwen 3**96 - URL: https://ollama.com/library/qwen39798- **GPT-OSS**99 - URL: https://ollama.com/library/gpt-oss100101- **DeepSeek-v3.1**102 - URL: https://ollama.com/library/deepseek-v3.1103104- **DeepSeek R1**105 - URL: https://ollama.com/library/deepseek-r1106107- **thinking models**108 - URL: https://ollama.com/search?c=thinking109110### Enable thinking in API calls111112### Stream the reasoning trace113114### CLI quick reference115116### Calling a single tool117118### Parallel tool calling119120### Multi-turn tool calling (Agent loop)121122### Tool calling with streaming123124### Using functions as tools with Ollama Python SDK125126### Quick start127128### Usage with Ollama's API129130### Authentication131132### Web search API133134### Web fetch API135136### Building a search agent137138### MCP Server139140### Cloud Models141142### Cloud API access143144### Local only145146### Setting context length147148### CPU only149150### Nvidia GPU151152- **Link**: Configure the repository153154```shell theme={"system"}155curl -fsSL \156 | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg157curl -fsSL https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list \158 | sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' \159 | sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list160sudo apt-get update161```162 - URL: https://nvidia.github.io/libnvidia-container/gpgkey163164- **Link**: Configure the repository165166```shell theme={"system"}167curl -fsSL \168 | sudo tee /etc/yum.repos.d/nvidia-container-toolkit.repo169```170 - URL: https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo171172### AMD GPU173174### Vulkan Support175176### Run model locally177178### Try different models179180### How can I upgrade Ollama?181182### How can I view the logs?183184### Is my GPU compatible with Ollama?185186### How can I specify the context window size?187188### How can I tell if my model was loaded onto the GPU?189190### How do I configure Ollama server?191192### How do I use Ollama behind a proxy?193194### Does Ollama send my prompts and answers back to ollama.com?195196### How do I disable Ollama's cloud features?197198### How can I expose Ollama on my network?199200### How can I use Ollama with a proxy server?201202### How can I use Ollama with ngrok?203204### How can I use Ollama with Cloudflare Tunnel?205206### How can I allow additional web origins to access Ollama?207208### Where are models stored?209210### How can I use Ollama in Visual Studio Code?211212### How do I use Ollama with GPU acceleration in Docker?213214### Why is networking slow in WSL2 on Windows 10?215216### How can I preload a model into Ollama to get faster response times?217218### How do I keep a model loaded in memory or make it unload immediately?219220### How do I manage the maximum number of requests the Ollama server can queue?221222### How does Ollama handle concurrent requests?223224### How does Ollama load models on multiple GPUs?225226### How can I enable Flash Attention?227228### How can I set the quantization type for the K/V cache?229230### Where can I find my Ollama Public Key?231232- **Ollama Cloud**233 - URL: https://ollama.com/cloud234235- **https://ollama.com/settings/keys**236 - URL: https://ollama.com/settings/keys237238### How can I stop Ollama from starting when I login to my computer?239240### Nvidia241242### AMD Radeon243244### Metal (Apple GPUs)245246### Vulkan GPU Support247248- **https://dgpu-docs.intel.com/driver/client/overview.html**249 - URL: https://dgpu-docs.intel.com/driver/client/overview.html250251- **https://amdgpu-install.readthedocs.io/en/latest/install-script.html#specifying-a-vulkan-implementation**252 - URL: https://amdgpu-install.readthedocs.io/en/latest/install-script.html#specifying-a-vulkan-implementation253254### Table of Contents255256- **Importing a Safetensors adapter**257 - URL: #Importing-a-fine-tuned-adapter-from-Safetensors-weights258259- **Importing a Safetensors model**260 - URL: #Importing-a-model-from-Safetensors-weights261262- **Importing a GGUF file**263 - URL: #Importing-a-GGUF-based-model-or-adapter264265- **Sharing models on ollama.com**266 - URL: #Sharing-your-model-on-ollamacom267268### Importing a fine tuned adapter from Safetensors weights269270- **fine tuning framework**271 - URL: https://huggingface.co/docs/transformers/en/training272273- **Unsloth**274 - URL: https://github.com/unslothai/unsloth275276- **MLX**277 - URL: https://github.com/ml-explore/mlx278279### Importing a model from Safetensors weights280281### Importing a GGUF based model or adapter282283### Quantizing a Model284285### Sharing your model on ollama.com286287### Libraries288289### Community290291### Install292293### Usage with Ollama294295### Recommended Models296297### Install298299### Usage with Ollama300301### Connecting to ollama.com302303- **API key**304 - URL: https://ollama.com/settings/keys305306- **Link**: Click on `Use custom base URL` and set it to `307 - URL: https://ollama.com`308309### Install310311### Usage with Ollama312313### Connecting to ollama.com314315### Install316317### Usage with Ollama318319### Cloud Models320321### Connecting to ollama.com322323- **API key**324 - URL: https://ollama.com/settings/keys325326- **Link**: Add the cloud configuration block to `~/.factory/config.json`:327328```json theme={"system"}329{330 "custom_models": [331 {332 "model_display_name": "qwen3-coder [Ollama Cloud]",333 "model": "qwen3-coder:480b",334 "base_url": "335 "api_key": "OLLAMA_API_KEY",336 "provider": "generic-chat-completion-api",337 "max_tokens": 128000338 }339 ]340}341```342 - URL: https://ollama.com/v1/",343344### Goose Desktop345346- **Link**: Confirm **API Host** is ` and click Submit347 - URL: http://localhost:11434`348349- **API key**350 - URL: https://ollama.com/settings/keys351352- **Link**: In Goose, set **API Host** to `353 - URL: https://ollama.com`354355### Goose CLI356357- **API key**358 - URL: https://ollama.com/settings/keys359360- **Link**: Update **OLLAMA\_HOST** to `361 - URL: https://ollama.com`362363### Coding Agents364365- **Claude Code**366 - URL: /integrations/claude-code367368- **Codex**369 - URL: /integrations/codex370371- **OpenCode**372 - URL: /integrations/opencode373374- **Droid**375 - URL: /integrations/droid376377- **Goose**378 - URL: /integrations/goose379380- **Pi**381 - URL: /integrations/pi382383### Assistants384385- **OpenClaw**386 - URL: /integrations/openclaw387388### IDEs & Editors389390- **VS Code**391 - URL: /integrations/vscode392393- **Cline**394 - URL: /integrations/cline395396- **Roo Code**397 - URL: /integrations/roo-code398399- **JetBrains**400 - URL: /integrations/jetbrains401402- **Xcode**403 - URL: /integrations/xcode404405- **Zed**406 - URL: /integrations/zed407408### Chat & RAG409410- **Onyx**411 - URL: /integrations/onyx412413### Automation414415- **n8n**416 - URL: /integrations/n8n417418### Notebooks419420- **marimo**421 - URL: /integrations/marimo422423### Install424425### Usage with Ollama426427- **Link**: Confirm the **Host URL** is ` then click **Ok**428 - URL: http://localhost:11434`,429430### Install431432### Usage with Ollama433434- **Link**: In marimo, go to the user settings and go to the AI tab. From here435you can find and configure Ollama as an AI provider. For local use you436would typically point the base url to `437 - URL: http://localhost:11434/v1`.438439### Connecting to ollama.com440441### Install442443### Using Ollama Locally444445- **Link**: Confirm Base URL is set to ` if running locally or `http://host.docker.internal:11434` if running through docker and click **Save**446 - URL: http://localhost:11434`447448### Connecting to ollama.com449450- **API key**451 - URL: https://ollama.com/settings/keys452453- **Link**: Set the **API URL** to `454 - URL: https://ollama.com`455456### Overview457458### Install Onyx459460### Usage with Ollama461462- **Link**: Provide your **Ollama API URL** and select your models.463<Note>If you're running Onyx in Docker, to access your computer's local network use ` instead of `http://127.0.0.1`.</Note>464 - URL: http://host.docker.internal`465466### Send your first query467468### Quick start469470### Configure without launching471472### Recommended models473474### Connect messaging apps475476### Stopping the gateway477478### Install479480### Usage with Ollama481482### Cloud Models483484### Connecting to ollama.com485486- **API key**487 - URL: https://ollama.com/settings/keys488489### Install490491### Usage with Ollama492493### Install494495### Usage with Ollama496497- **Link**: (Optional) Update `Base URL` if your Ollama instance is running remotely. The default is `498 - URL: http://localhost:11434`499500### Connecting to ollama.com501502- **API key**503 - URL: https://ollama.com/settings/keys504505- **Link**: Enable `Use custom base URL` and set it to `506 - URL: https://ollama.com`507508### Install509510### Usage with Ollama511512### Install513514### Usage with Ollama515516### Connecting to ollama.com directly517518- **API key**519 - URL: https://ollama.com/settings/keys520521- **Link**: Select **Internet Hosted** and enter URL as `522 - URL: https://ollama.com`523524### Install525526### Usage with Ollama527528- **Link**: Confirm the **Host URL** is ` then click **Connect**529 - URL: http://localhost:11434`,530531### Connecting to ollama.com532533- **API key**534 - URL: https://ollama.com/settings/keys535536- **Link**: Set the **API URL** to `537 - URL: https://ollama.com`538539### Install540541### Manual install542543### Customizing544545### Updating546547### Installing specific versions548549### Viewing logs550551### Uninstall552553### System Requirements554555### Filesystem Requirements556557### Troubleshooting558559### Uninstall560561### Table of Contents562563- **Format**564 - URL: #format565566- **Examples**567 - URL: #examples568569- **Instructions**570 - URL: #instructions571572- **Notes**573 - URL: #notes574575### Format576577### Examples578579### Instructions580581### Notes582583### Get Started584585### Assistants586587### Coding588589### API590591### LLM libraries592593### Installing older or pre-release versions on Linux594595### Linux tmp noexec596597### Linux docker598599### NVIDIA GPU Discovery600601### AMD GPU Discovery602603### Multiple AMD GPUs604605- **https://rocm.docs.amd.com/projects/radeon/en/latest/docs/install/native\_linux/mgpu.html#mgpu-known-issues-and-limitations**606 - URL: https://rocm.docs.amd.com/projects/radeon/en/latest/docs/install/native_linux/mgpu.html#mgpu-known-issues-and-limitations607608### Windows Terminal Errors609610### System Requirements611612- **https://www.amd.com/en/support**613 - URL: https://www.amd.com/en/support614615### Filesystem Requirements616617### API Access618619### Troubleshooting620621### Uninstall622623### Standalone CLI624625## How to Use This Skill626627Reference these resources when working with Windows.
Run npx skillmds@latest add diegosouzapw/windows-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
Windows It is listed under Coding & Dev Tools on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: WARNING, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
diegosouzapw (@diegosouzapw) published this skill. Their other Agent Skills are listed on their SkillMD profile.
diegosouzapw