Blog
Guides, comparisons, and engineering notes from the team building AnyRouter — one OpenAI-compatible endpoint in front of every model.
Changelog
New models added to the catalog — frontier releases as they land.
- Nov 9, 2026Ollama models over RelayQwen 2.5 Coder 0.5B and Mistral 7B Instruct v0.3 join the catalog as on-device listings served by your own Ollama over AnyRouter Relay at $0 — and the existing Qwen 2.5 Coder 32B gains the same Ollama hop.Qwen 2.5 Coder 0.5BMistral 7B Instruct v0.3Qwen 2.5 Coder 32B
- Nov 8, 2026Ampersand v0.1humans&'s collaborator model joins the catalog — trained from GLM 5.3 to track what each person wants, flag conflicts, and share context in group chats, with a 1M-token context — BYOK through Vercel AI Gateway.Ampersand v0.1Hy4Decisions Fast
- Nov 7, 2026Inkling 2 SmallThinking Machines' second-generation open-weights model joins the catalog — 594B MoE, 17B active, vision input, up to 1M context — BYOK through Tinker's OpenAI-compatible API.Inkling 2 Small
Guides
How to route every model through one OpenAI-compatible endpoint — setup, coding agents, MCP, free models, and BYOK.
- Use AnyRouter in DeepSeek Harness
- Run Claude Code on free models in two lines
- Claude Code on anyrouter.dev
- Use OpenCode with AnyRouter — CLI launcher or opencode.json
- Run Claude Code on AnyRouter — wrapper script or CLI
- Sign in with AnyRouter: let users bring their own LLM account
- Run the Claude Agent SDK on any model
- Use AnyRouter with the Vercel AI SDK
- Get a free NVIDIA API key and route it through one URL
- Set up a Hermes agent with AnyRouter
- Point Claude Code at a custom base URL
- Use the OpenAI SDK with any model
- Cheapest LLM API: how to actually shop for cost per token
- Free LLM API access for students and side projects
- MCP servers for AI agents: what they are and how to connect one
- One API for every LLM
- BYOK: bring your own keys to every provider
- Run coding agents through one gateway
- An MCP gateway with managed server access
- A unified AI API: one endpoint, every provider dialect
- The top website for free LLM models — and how they're actually free
- Free models, just $2 a month
Comparisons
How AnyRouter stacks up against calling providers directly or through another gateway — pricing, markup, and lock-in.
Product
New features, launches, and changes to the gateway — credits, the shared key pool, on-device models, and the dashboard.
- AnyRouter turns 3 months, in numbers
- CLI update: a separate key for each coding agent
- Introducing AnyRouter Artifacts: publish a page, get a link
- What is new: October 2026
- MiniMax M3.1 Flash is free on Go and higher
- Free tokens from a pool of idle keys
- Introducing the AnyRouter Pool
- How we protect your provider keys
- New models, 20 Sep – 10 Oct 2026
- Where every -latest alias points today
- Six decision models, two of them free
- Latest aliases: name a family, always get a live model
- anyr decision puts structured decisions in your terminal
- Jev Decisions is now on its model page
- Space Bunny Alpha is free on AnyRouter
- A shared key pool, and a free-token pool on the other side
- Donate a key for Go — no card
- AnyRouter pricing: Free vs Go (the $1 posts are wrong)
- Ox Alpha is now GLM-5.3-Flash
- Qwen3.8-27B and DeepSeek V4 Flash, 50% off
- CLI update: multiple accounts, terminal key management, and agents that delegate
- Bring your CommandCode credential to AnyRouter
- Bring your Antigravity credential to AnyRouter
- Apple's on-device Foundation Model, now on the AnyRouter API
- Go on AnyRouter for $2 — or free if you donate a key
- Donate a key to unblock the Pool of free tokens
- Bring your own keys — every provider behind one API
- A clearer way to pay: Free, Standard, and pay-as-you-go
Data
What the traffic actually shows — most-used models, real latency, and the economics of running a router.
Engineering
How the gateway is built on Cloudflare Workers — routing, failover, streaming, and the edge trade-offs behind it.
- How we cut our D1 bill by moving request logs to Cloudflare Basin
- How our CI/CD changed when coding agents took over
- How we sped up our local CI without dropping a check
- How we made AnyRouter faster
- How agents deploy AnyRouter: one Worker at a time
- How agents ship AnyRouter: the merge and deploy loop
- How automatic failover works in an LLM gateway
- Building an LLM gateway on Cloudflare Workers