Millwright
Rust-based, self-hosted LLM router
Details
- External ID
- 49011806
- Source
- HN
- Company
- —
- Product
- Millwright
- Website domain
- github.com
- Launched
- July 22, 2026
- Cohort
- —
- Upvotes
- 10
- Upvotes percentile
- 0.5418160095579451
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Hey HN,With the news of OpenRouter possibly being acquired and proliferation of hosted LLM routers (i.e. Ramp Router, Vercel’s AI Gateway), I saw the need for a self hosted solution focused on cost savings, transparency, and performance. So, I built an open sourced router with a simple CLI interface that can easily sit between coding agents and GenAI workloads.For the curious and lazy, at the moment, Millwright has the tools for,- Providers: OpenAI-compatible APIs, Anthropic, Amazon Bedrock- Routing: policy-controlled model roles (cheap, mid, frontier), cheapest healthy route selection- Protocols: OpenAI Chat Completions, Anthropic Messages, text and tool translation- Cache Affinity: role-scoped session lanes without serializing concurrent agent traffic- Spend Tracking: per-team costs, cache usage, model/provider mix, request traces- Cost Analysis: measured usage and modeled candidate economics (HTML, Markdown, JSON)- Reliability: bounded failover, circuit breakers, timeouts, concurrency limits- Setup: interactive provider, model, and pricing configuration without storing provider secrets- Deployment: one Rust binary, Docker, SQLite or PostgreSQLFull disclosure: parts of the codebase were built with AI coding agents. All feedback is welcome, I’d especially value feedback on the routing policy, provider coverage, and anything that would block you from self-hosting it. Feel free to open feature/request and/or contribute as well.
Enrichment
- Theme
- ML inference and model optimization
- Vertical
- Horizontal
- Function
- Model & infra
- Audience
- Developer
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- self-hosted llm routing layer
- Manually corrected
- False
Could you build this?
Partial The basic proxying and routing logic is vibe-codeable, but building a production-grade, low-latency, resilient LLM gateway in Rust with high-throughput streaming, fallbacks, and connection pooling requires solid systems engineering.
What it would actually take: The software is built with Rust using Tokio, Axum/Hyper, and Tower services to handle streaming HTTP SSE connections, token-bucket rate limiting, semantic caching, and dynamic failovers across heterogeneous provider APIs. The hard parts are maintaining near-zero proxy latency overhead under heavy concurrency, handling divergent provider token-counting conventions, and implementing reliable circuit-breaker patterns.
Discussion
7 comments analyzed.
Competitors mentioned: role-model (on-device router alternative)
Concerns raised: Energy consumption/environmental impact, Vendor lock-in risk with foreign companies, Cognitive dependency - losing ability to use own brain, Pricing will increase after undercharging phase, Legal uncertainty around training on copyrighted software
Competitors
Other products that read as similar to this one — 254 launches clear the similarity bar, closest 8 shown.
Attention rank: #123 of 255 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 266 days after the earliest competitor.
- We built open OpenRouter that turns usage into a better model · hn · 2026-08-27 · 222 upvotes · similarity 0.52
- LLM-Gateway · hn · 2026-03-27 · 7 upvotes · similarity 0.51
- Foreman, a self-hosted LLM gateway for cost aware model routing · hn · 2026-07-08 · 15 upvotes · similarity 0.50
- Smart model routing directly in Claude, Codex and Cursor · hn · 2026-06-26 · 216 upvotes · similarity 0.49
- TheLittleHost · hn · 2026-03-16 · 7 upvotes · similarity 0.49
- LLMRouter · hn · 2025-12-31 · 5 upvotes · similarity 0.47
- ClawRouter · hn · 2026-02-05 · 12 upvotes · similarity 0.46
- I built managed OpenClaw hosting with 60s provisioning in 6 days · hn · 2026-02-11 · 9 upvotes · similarity 0.46
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a model & infra tool for Fintech yet.