Conifer
Local-first least cost routing system to reduce 80%+ token spend
About Conifer
Conifer is one app that connects you to every AI model (local and cloud) and automatically routes each request to the cheapest model that accomplishes your goal. Most requests run free on your own hardware, so teams cut their AI bill by ~80% while replacing a dozen provider contracts with one interface and one invoice. For sensitive material like financials or user data, a secure mode pins everything on-device with no external API calls.
Public traction evidence
Each signal links to the public source used for attribution.
- GitHub
ConiferKit/sage
ConiferKit/sage. YC S26 snapshot lists Conifer with official GitHub org https://github.com/ConiferKit; repo lives under that org.
- X
Conifer routes 80 percent of AI requests to local hardware
Conifer routes 80 percent of AI requests to local hardware
- X
Excited to share that @coniferbuild is part of the @ycombinator S26 batch. Looking forward to getting to work with @gustaf ! Token costs are eye-watering. @charles_v11 and I burned through $13,000 in Claude credits in just five days, and that's a fraction of what every company Show more
Excited to share that @coniferbuild is part of the @ycombinator S26 batch. Looking forward to getting to work with @gustaf ! Token costs are eye-watering. @charles_v11 and I burned through $13,000 in Claude credits in just five days, and that's a fraction of what every company...
- X
Fresh essay on scaling laws. Sara Hooker spent years inside DeepMind and Cohere watching scaling laws get treated as gospel. Then she wr...
Fresh essay on scaling laws. Sara Hooker spent years inside DeepMind and Cohere watching scaling laws get treated as gospel. Then she wrote down the heresy: Falcon 180B was state of the art in 2023. One year later it lost to a model 22x smaller. The $700B question isn't whether...
- X
Kimi beats Fable and Sol in Code Arena!!
Kimi beats Fable and Sol in Code Arena!!
- X
So many great points, but the error-correction piece is the one people will underrate. LeCun's (1−ε)ⁿ doom assumes per-step errors are ii...
So many great points, but the error-correction piece is the one people will underrate. LeCun's (1−ε)ⁿ doom assumes per-step errors are iid and absorbing, which they're not. Trained agents learn a restoring force back so long horizon behavior looks like a mean reverting walk. 👏
- X
Pretty frustrating using Claude this month. Models seem slower, constantly hitting guardrails, and burning tokens. Planning is still supe...
Pretty frustrating using Claude this month. Models seem slower, constantly hitting guardrails, and burning tokens. Planning is still superior, but definitely stresses the need for multi model setups
- X
Kimi K3 Benchmarks - seriously where is for profit in 18 months
Kimi K3 Benchmarks - seriously where is for profit in 18 months