VoidLLM BlogTechnical articles from the VoidLLM team on LLM proxies, performance, privacy, and open source.https://voidllm.ai/Writing MCP tool descriptions that LLMs actually followhttps://voidllm.ai/blog/tool-descriptions-llms-actually-follow/https://voidllm.ai/blog/tool-descriptions-llms-actually-follow/An LLM that ignores your tool description costs round-trips. We rewrote the Code Mode execute_code description with STRONG PREFERENCE labels, workflow steps, and concrete code patterns. Here's what changed and why.Tue, 05 May 2026 00:00:00 GMTHow Code Mode learns what tools returnhttps://voidllm.ai/blog/code-mode-learns-tool-return-types/https://voidllm.ai/blog/code-mode-learns-tool-return-types/MCP tools advertise inputs but not outputs. We taught Code Mode to learn return types from the first successful call and surface them as TypeScript on the next discovery.Thu, 30 Apr 2026 00:00:00 GMTWhy VoidMCP is its own projecthttps://voidllm.ai/blog/why-voidmcp-is-its-own-project/https://voidllm.ai/blog/why-voidmcp-is-its-own-project/Code Mode is great, but you shouldn't need an LLM proxy to use it. VoidMCP is a separate Go binary that brings Code Mode and the MCP gateway to anyone who just wants their IDE to talk to multiple MCP servers.Sat, 25 Apr 2026 00:00:00 GMTMigrating from LiteLLM to VoidLLM: A Practical Guidehttps://voidllm.ai/blog/migrating-from-litellm-to-voidllm/https://voidllm.ai/blog/migrating-from-litellm-to-voidllm/Step by step migration from LiteLLM to VoidLLM - config translation, key rollout, client switchover, and gotchas to know before you start.Sun, 19 Apr 2026 00:00:00 GMTHow to Allocate LLM Costs Across Teams and Departmentshttps://voidllm.ai/blog/llm-cost-allocation-teams/https://voidllm.ai/blog/llm-cost-allocation-teams/A practical guide to tracking and allocating LLM spend across teams using org/team/user/key hierarchies. No more monolithic OpenAI bills.Fri, 17 Apr 2026 00:00:00 GMTVoidLLM vs OpenRouter: Self-Hosted Proxy vs Hosted Aggregatorhttps://voidllm.ai/blog/voidllm-vs-openrouter/https://voidllm.ai/blog/voidllm-vs-openrouter/An honest comparison between VoidLLM (self-hosted LLM proxy) and OpenRouter (hosted LLM aggregator) - when each makes sense.Sat, 11 Apr 2026 00:00:00 GMTEU AI Act and LLM Proxies - Where VoidLLM Standshttps://voidllm.ai/blog/eu-ai-act-llm-proxies-where-voidllm-stands/https://voidllm.ai/blog/eu-ai-act-llm-proxies-where-voidllm-stands/Most LLM proxies log your prompts. The EU AI Act makes that a compliance problem. Here's how VoidLLM's architecture simplifies things.Wed, 01 Apr 2026 00:00:00 GMTVoidLLM vs LiteLLM: Honest Comparisonhttps://voidllm.ai/blog/voidllm-vs-litellm/https://voidllm.ai/blog/voidllm-vs-litellm/An honest look at how VoidLLM and LiteLLM differ - where each one shines and where it doesn't.Wed, 01 Apr 2026 00:00:00 GMTLoad Balancing and Failover Across LLM Providershttps://voidllm.ai/blog/load-balancing-failover-llm-providers/https://voidllm.ai/blog/load-balancing-failover-llm-providers/Route LLM requests across multiple deployments with automatic failover, health-aware routing, and four balancing strategies.Tue, 31 Mar 2026 00:00:00 GMTCode Mode: Let AI Agents Write Scripts, Not Chathttps://voidllm.ai/blog/code-mode-ai-agents-write-scripts/https://voidllm.ai/blog/code-mode-ai-agents-write-scripts/VoidLLM's Code Mode lets AI agents orchestrate multiple MCP tool calls in a single WASM-sandboxed JavaScript execution. No round-trips, no latency penalty.Mon, 30 Mar 2026 00:00:00 GMTProxy Overhead Benchmarks: How Fast is VoidLLM?https://voidllm.ai/blog/proxy-overhead-benchmarks/https://voidllm.ai/blog/proxy-overhead-benchmarks/Real benchmark numbers for VoidLLM's LLM proxy, MCP gateway, and Code Mode. Sub-500 microsecond overhead at 1000 requests per second.Sun, 29 Mar 2026 00:00:00 GMTMCP Gateway: Give Your AI Tools a Single Front Doorhttps://voidllm.ai/blog/mcp-gateway-single-front-door/https://voidllm.ai/blog/mcp-gateway-single-front-door/VoidLLM acts as an MCP gateway - proxy, manage, and control access to external MCP servers from one place.Sat, 28 Mar 2026 00:00:00 GMTZero-Knowledge by Architecture, Not by Policyhttps://voidllm.ai/blog/zero-knowledge-architecture/https://voidllm.ai/blog/zero-knowledge-architecture/How VoidLLM enforces privacy at the code level - no content logging, no opt-out, no exceptions.Fri, 27 Mar 2026 00:00:00 GMTDrop-In Replacement for the OpenAI SDKhttps://voidllm.ai/blog/drop-in-replacement-openai-sdk/https://voidllm.ai/blog/drop-in-replacement-openai-sdk/Switch from direct OpenAI API calls to VoidLLM by changing one line. Route to any provider - Anthropic, Azure, vLLM - through the same SDK.Thu, 26 Mar 2026 00:00:00 GMTGetting Started: Docker to First Request in 3 Minuteshttps://voidllm.ai/blog/getting-started-docker-to-first-request/https://voidllm.ai/blog/getting-started-docker-to-first-request/A quick walkthrough of VoidLLM - from docker run to your first proxied LLM request, with a look at the built-in UI.Wed, 25 Mar 2026 00:00:00 GMTConnect Cursor, Windsurf, and Claude Code to VoidLLMhttps://voidllm.ai/blog/connect-cursor-windsurf-claude-code/https://voidllm.ai/blog/connect-cursor-windsurf-claude-code/Step-by-step setup for using VoidLLM as your LLM proxy in Cursor and Windsurf, and as an MCP server in Claude Code.Tue, 24 Mar 2026 00:00:00 GMTWhy We Built VoidLLMhttps://voidllm.ai/blog/why-we-built-voidllm/https://voidllm.ai/blog/why-we-built-voidllm/Existing LLM gateways are bloated, cloud-only, or charge per request. We built a self-hosted alternative that doesn't compromise on privacy or performance.Sun, 22 Mar 2026 00:00:00 GMT