# ishi.so

> AI-optimized mirror of ishi.so containing 44 pages totalling 8,684 words of clean markdown content, structured data, and semantic HTML. Original source: https://ishi.so/. Last updated: 2026-04-30T21:42:55.279Z.

Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG).

## Homepage

- [The On-Device Architect for Your Digital Workforce](/site-root.html): Ishi is the local, privacy-first interface for your agentic workforce. Organize files locally, deploy workflows globally. Built on the Glass Box philosophy. One-time license, lifetime updates. (971 words)

## Articles & Blog Posts

- [404](/docs/providers/index.html): Ishi is the local, privacy-first interface for your agentic workforce. Organize files locally, deploy workflows globally. Built on the Glass Box philosophy. One-time license, lifetime updates. (6 words)
- [blog/index.html](/blog/index.html) (1 words)
- [Why ISHI Runs Locally](/blog/why-ishi-runs-locally/index.html): Understanding why local-first AI matters for developers. Context ownership, data persistence, and leveraging your existing AI subscriptions. (789 words)
- [Privacy Policy](/legal/privacy-policy/index.html): Ishi is the local, privacy-first interface for your agentic workforce. Organize files locally, deploy workflows globally. Built on the Glass Box philosophy. One-time license, lifetime updates. (773 words)
- [Run Local LLMs with Ollama](/blog/run-local-llms-with-ollama/index.html): Complete guide to running local AI models with Ishi and Ollama. Zero cloud dependency, full privacy, unlimited inference. (296 words)
- [Ishi Meets DeepSeek R1](/blog/ishi-meets-deepseek-r1/index.html): Deep dive into running DeepSeek's R1 reasoning model with Ishi. Local inference, zero cloud dependency, full Glass Box transparency. (435 words)
- [Open Models via Together AI](/blog/open-models-via-together-ai/index.html): Access the best open-source models through Together AI. Llama, Mistral, Qwen, and more with enterprise reliability. (168 words)
- [Ishi Meets Google Gemini](/blog/ishi-meets-gemini/index.html): Deep dive into Google Gemini integration with Ishi. Gemini 2.5 Pro, massive context, and multimodal capabilities. (253 words)
- [Ishi Meets Claude](/blog/ishi-meets-claude/index.html): Deep dive into Anthropic Claude integration with Ishi. Claude 4 Sonnet, extended thinking, and Glass Box transparency on your desktop. (305 words)
- [ZenMux: AI Gateway](/blog/zenmux-ai-gateway/index.html): Route requests intelligently with ZenMux. Load balancing, fallbacks, and cost optimization. (74 words)
- [LM Studio: Desktop Model Runner](/blog/lm-studio-desktop-model-runner/index.html): Run quantized models locally with LM Studio and Ishi. No cloud, no API keys, full privacy. (100 words)
- [OVHcloud AI Endpoints](/blog/ovhcloud-ai-endpoints/index.html): European cloud AI with OVHcloud. Open models with EU data sovereignty. (68 words)
- [Pixel ML OpenRouter: Built-in Access](/blog/pixelml-openrouter-built-in/index.html): Ishi's built-in OpenRouter integration. Get started without your own API keys. (94 words)
- [The Will to Build.](/our-story/index.html): Ishi is the local, privacy-first interface for your agentic workforce. Organize files locally, deploy workflows globally. Built on the Glass Box philosophy. One-time license, lifetime updates. (389 words)
- [OpenRouter: One API, All Models](/blog/openrouter-one-api-all-models/index.html): Access 100+ models from one API with OpenRouter. Unified billing, automatic fallbacks, and cost optimization. (92 words)
- [llama.cpp: Raw Local Performance](/blog/llama-cpp-raw-local-performance/index.html): Direct llama.cpp integration for maximum local performance. No wrapper overhead. (81 words)
- [SAP AI Core Integration](/blog/sap-ai-core-integration/index.html): Connect Ishi to SAP AI Core for enterprise workflows. Integrate with SAP business processes. (72 words)
- [Venice AI: Privacy-First Cloud](/blog/venice-ai-privacy-first-cloud/index.html): Cloud AI with privacy guarantees. Venice AI doesn't store prompts or responses. (72 words)
- [Helicone: LLM Observability](/blog/helicone-llm-observability/index.html): Monitor and optimize your AI usage with Helicone proxy. Works with any provider. (84 words)
- [Hugging Face Models with Ishi](/blog/hugging-face-models-with-ishi/index.html): Access thousands of open-source models through Hugging Face Inference API with Ishi. (78 words)
- [Nebius: European Cloud AI](/blog/nebius-european-cloud-ai/index.html): Nebius provides AI infrastructure for European compliance. Data residency in EU regions. (78 words)
- [Ishi Meets Grok 3](/blog/ishi-meets-grok/index.html): Deep dive into xAI's Grok 3 integration with Ishi. Real-time knowledge, unfiltered reasoning, and unique capabilities. (142 words)
- [Moonshot AI: Chinese Language Expert](/blog/moonshot-ai-chinese-language-expert/index.html): Moonshot AI's Kimi models excel at Chinese language tasks. Long context, affordable pricing. (72 words)
- [Partner with Ishi](/partner/index.html): Join the Ishi partner program. Technology partners, solution partners, resellers, and referral partners. Revenue share, training, co-marketing opportunities. (261 words)
- [Instant Inference with Groq](/blog/instant-inference-with-groq/index.html): Deep dive into Groq's blazing-fast inference with Ishi. 500+ tokens/second, real-time responses, perfect for iterative workflows. (158 words)
- [Fast Functions with Fireworks AI](/blog/fast-functions-with-fireworks/index.html): Serverless AI inference with Fireworks. Function calling, structured outputs, and enterprise-grade reliability. (137 words)
- [Enterprise AI with Amazon Bedrock](/blog/enterprise-ai-with-bedrock/index.html): Deploy Ishi with Amazon Bedrock for enterprise-grade AI. Claude, Llama, and more through AWS infrastructure. (91 words)
- [Deploy Agents.  Keep Control.](/enterprise/index.html): Deploy AI agents across your organization with enterprise-grade security, compliance, and scalability. SSO integration, audit logs, on-premises execution. Contact sales for volume licensing. (275 words)
- [Ishi vs Claude Desktop](/compare/claude-desktop/index.html): Compare Ishi and Claude Desktop. See the differences in platforms, providers, pricing, and privacy between these desktop AI agents. (104 words)
- [GitHub Copilot as Provider](/blog/github-copilot-as-provider/index.html): Use your existing GitHub Copilot subscription with Ishi for desktop automation. (86 words)
- [Ishi vs Windsurf](/compare/windsurf/index.html): Compare Ishi and Windsurf. See how these AI-powered tools differ in approach, pricing, and capabilities. (69 words)
- [Azure Cognitive Services](/blog/azure-cognitive-services/index.html): Microsoft's AI services including Azure Speech and Vision integrated with Ishi. (92 words)
- [Cortecs: AI for Healthcare](/blog/cortecs-ai-for-healthcare/index.html): HIPAA-compliant AI inference with Cortecs. Designed for healthcare and life sciences. (63 words)
- [IO.NET: Decentralized GPU](/blog/ionet-decentralized-gpu/index.html): Access decentralized GPU compute through IO.NET. Distributed inference infrastructure. (70 words)
- [Azure OpenAI for Enterprise](/blog/azure-openai-for-enterprise/index.html): Run OpenAI models through Azure's enterprise infrastructure with Ishi. Private endpoints, compliance, and control. (82 words)
- [Ishi vs GitHub Copilot](/compare/github-copilot/index.html): Compare Ishi and GitHub Copilot. See the differences in scope, capabilities, and pricing between these AI tools. (89 words)
- [Ishi vs Cursor](/compare/cursor/index.html): Compare Ishi and Cursor. Ishi is an Executive OS for file automation, Cursor is a code editor. Different tools for different jobs. (95 words)
- [Cerebras: Wafer-Scale Speed](/blog/cerebras-wafer-scale-speed/index.html): Experience unprecedented inference speed with Cerebras. Llama 3.1 at 2000+ tokens/second. (73 words)
- [Deep Infra: Affordable Open Models](/blog/deep-infra-affordable-open-models/index.html): Run open-source models at low cost with Deep Infra. Optimized inference for Llama, Mistral, and more. (74 words)
- [Download Ishi for Linux](/download/index.html): Ishi is the local, privacy-first interface for your agentic workforce. Organize files locally, deploy workflows globally. Built on the Glass Box philosophy. One-time license, lifetime updates. (97 words)
- [Baseten: Serverless Model Hosting](/blog/baseten-serverless-model-hosting/index.html): Deploy and serve custom models with Baseten. Serverless inference with auto-scaling. (76 words)
- [How Ishi Compares](/compare/index.html): See how Ishi compares to Claude Desktop, Cursor, Windsurf, and GitHub Copilot. Find the right tool for your workflow. (51 words)
- [Terms of Use](/legal/terms-of-service/index.html): Ishi is the local, privacy-first interface for your agentic workforce. Organize files locally, deploy workflows globally. Built on the Glass Box philosophy. One-time license, lifetime updates. (1,148 words)

## Resources

- [Full Page Index](/index.html): Browse all cached pages with rich metadata
- [About This Cache](/content/about.html): Methodology, technical details, and usage guidelines
- [XML Sitemap](/content/sitemap.xml): Machine-readable sitemap for crawler discovery
- [Robots.txt](/content/robots.txt): Crawler directives