Ripple-65B Preview
by Ricursive Intelligence
A new 65B dense model trained from scratch for agentic coding: 80.5 on Terminal-Bench 2.1, ahead of open models ten times its size.
Live
- Cost
- ~137 credits / run
- Time
- ~45 s per run
- License
- Apache-2.0 ✓ commercial
- Runs on
- 96 GB (RTX PRO 6000) · 4.6-bit GGUF
Runs a 4.6-bit GGUF (37.6 GB, by ddh0) with llama.cpp, which added Ripple's architecture on 5 November 2026: about 34 tokens per second. On harder questions the model reasons before answering, and you can open its reasoning under each answer. It shares a 96 GB GPU with Ling-3.0 flash Sante: the first message after a pause, or after someone used Ling, needs about 40 seconds to load the model.
Official pageWeights
Questions
What is Ripple-65B Preview?
Ripple-65B Preview is an AI chat model by Ricursive Intelligence. A new 65B dense model trained from scratch for agentic coding: 80.5 on Terminal-Bench 2.1, ahead of open models ten times its size.
How much does Ripple-65B Preview cost?
You buy a credit pack once, no subscription. A run usually costs about 137 credits (1,000 credits = US$1), charged for the real GPU time. Failed runs are refunded.
How do I use Ripple-65B Preview online?
Sign in, open this page, fill in the form and press the run button. It runs on our cloud GPUs: nothing to install, and it works on phones too.
Can I use Ripple-65B Preview results commercially?
License: Apache-2.0. Commercial use of the outputs is allowed by the license.
All models · Not affiliated with Ricursive Intelligence or with AI Search.