Conifer: Local-first least cost routing system
Stop paying cloud prices for every token. We route 80% of requests to your hardware at no cost
Details
- External ID
- 105897
- Source
- YC
- Company
- Conifer
- Product
- Conifer: Local-first least cost routing system
- Website domain
- conifer.build
- Launched
- July 15, 2026
- Cohort
- Summer 2026
- Upvotes
- 10
- Upvotes percentile
- 0.4016393442622951
- Tags
- Artificial Intelligence, Developer Tools, B2B, Security
- Fetched at
- Oct. 1, 2026, 1 a.m.
- Updated at
- Oct. 1, 2026, 1 a.m.
Enrichment
- Theme
- payments, billing, and wallet infrastructure
- Vertical
- Horizontal
- Function
- Model & infra
- Audience
- Developer
- AI stance
- AI feature
- Project type
- Commercial product
- Normalized one-liner
- local routing to reduce token costs
- Manually corrected
- False
Could you build this?
Partial Building a client-side proxy that forwards easy queries to a local model (via Ollama or llama.cpp) and fallbacks to cloud APIs is vibe-codeable, but achieving automated, reliable semantic complexity routing with minimal latency overhead is technically non-trivial.
What it would actually take: A real solution uses a lightweight local proxy daemon (written in Go or Rust) exposing an OpenAI-compatible endpoint that inspects incoming prompt tokens, assesses semantic complexity and context size via lightweight classifier heads (e.g. small DeBERTa/BERT or embedding probes), and dynamically dispatches either to a local llama.cpp/vLLM instance or cloud providers. The hardest component is low-latency intent/difficulty scoring (sub-10ms) that doesn't degrade output quality or introduce false-positive routings. Systems-level optimization and local hardware capability profiling across various OSs and GPUs are necessary.
Competitors
Other products that read as similar to this one — 118 launches clear the similarity bar, closest 8 shown.
Attention rank: #72 of 119 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 252 days after the earliest competitor.
- Trading on-chain with 0 gas: A <1ms DEX router at 0.50% fees · hn · 2026-09-17 · 9 upvotes · similarity 0.49
- genpark-token-bucket-rate-limiter-burst-controller-skill · github · 2026-09-29 · 7 upvotes · similarity 0.45
- Multichain-Drainer-Tool · github · 2026-09-10 · 30 upvotes · similarity 0.43
- Unifold – Universal deposit infrastructure for on-chain apps · yc · 2026-02-17 · 139 upvotes · similarity 0.43
- ComptyAI · github · 2026-09-21 · 51 upvotes · similarity 0.43
- genpark-adaptive-token-rate-limiter-skill · github · 2026-09-14 · 8 upvotes · similarity 0.43
- genpark-adaptive-token-rate-limiter-skill · github · 2026-09-14 · 7 upvotes · similarity 0.43
- WavexAI · hn · 2026-09-22 · 6 upvotes · similarity 0.42
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.