Ricursive releases Ripple-65B open model scoring 80.5 on Terminal-Bench
First merged summary.
Ricursive released Ripple-65B Preview, a 65B dense open-weight model scoring about 80.5 on Terminal-Bench 2.1.
Ricursive has released Ripple-65B Preview, a dense 65-billion-parameter open-weight language model. The Decoder reports it was pretrained on 12 trillion tokens at a development cost of $32 million, scoring 80.52 on Terminal-Bench 2.1 and 79.1% on SWE-Bench Pro v2.0 Hard. MarkTechPost reports a Terminal-Bench 2.1 score of 80.5 and names the company Ricursive AI; The Decoder calls the developer Ricursive Intelligence. According to The Decoder, the weights are on Hugging Face under Apache 2.0 and optimized for Nvidia hardware.
- Dense Transformer with 65 billion parameters
- Pretrained on 12 trillion tokens, The Decoder reports
- Development cost $32 million, according to The Decoder
- Terminal-Bench 2.1: 80.52 (The Decoder); 80.5 (MarkTechPost)
- SWE-Bench Pro v2.0 Hard score 79.1%, The Decoder reports
- Apache 2.0 weights on Hugging Face; Nvidia-optimized
What each outlet adds
beyond the shared facts- MarkTechPostNames Ricursive AI and reports an 80.5 Terminal-Bench 2.1 score.
- The DecoderAdds token count, cost, license, SWE-Bench, and Nvidia optimization.
- AI models
- Ripple-65B PreviewRipple-65B
Coverage
oldest first · one row per article · UTC- · 6d ago
MarkTechPostFirst report
Ricursive AI releases Ripple-65B preview scoring 80.5 on Terminal-Bench 2.1
Ricursive AI releases Ripple-65B, a 65B dense open-weight model scoring 80.5 on Terminal-Bench 2.1.
- · 5d ago
The Decoder
Ricursive releases Ripple-65B Preview open model
Pretrained on 12 trillion tokens at a cost of $32 million, the model scores 80.52 on TerminalBench 2.1.