OneTriangle
1 known investors
Find your way into OneTriangle
OneTriangle provides an inference optimization service that transfers KV cache between models — running prefill on a small model and decode on a large one — to reduce large language model inference costs and latency by around 20%. It also hosts and serves open-weight models like Llama, Qwen, Mistral, Gemma, and DeepSeek as a managed, private setup for teams.
Investors · 1
Frequently asked questions
- What does OneTriangle do?
- OneTriangle provides an inference optimization service that transfers KV cache between models — running prefill on a small model and decode on a large one — to reduce large language model inference costs and latency by around 20%. It also hosts and serves open-weight models like Llama, Qwen, Mistral, Gemma, and DeepSeek as a managed, private setup for teams.
- Who are OneTriangle's investors?
- OneTriangle's investors include Vela Partners.