Skip to content
Live
Trending storyPast story

Nanbeige4.3-33B released

5 reports4 outlets12 days agoUpdate

The story so far

Developments

2 developments

  1. 5 Nov, 02:53 · 4 reports
    Ripple-65B Preview released
    The Decoder:Ripple-65B Preview delivers strong agentic coding results as a fully Apache 2.0 open US model
  2. 26 Oct, 05:15 · 1 reports
    Nanbeige4.3-33B released
    MarkTechPost:Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model

Coverage timeline

Follow the coverage to see every side of the story.

6 Nov
  1. ITHome · AI
    Ricursive Intelligence releases Ripple-65B Preview, stating it has the best agent programming performance in the US open weight model.

    Ricursive Intelligence releases Ripple-65B Preview model, claiming it has the best agent programming performance in the US open weight model.

5 Nov
  1. MarkTechPost
    Ricursive AI Releases Ripple-65B Preview: A 65B Dense Open-Weight Model Scoring 80.5 on Terminal-Bench 2.1

    Ricursive AI has released Ripple-65B Preview, a 65B dense open-source model, scoring 80.5 on Terminal-Bench 2.1, becoming the highest-scoring open-source model in the United States. Its parameter size is 65B (active parameters), pre-training used 12T tokens, with a maximum context length of 1M, and is released under the Apache 2.0 license.

  2. Hacker News · 300+
    Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding

    Ripple-65B Improves Computational Efficiency by Optimizing SFT and RL Schemes, Using Synthetic Data Pipelines and Quality Filtering, Employing a Hybrid RL Scheme with Truncated Importance Sampling, Score Centering, and Linear Length Penalty in a High-Quality Long-Term Synthetic Environment, Achieving a TerminalBench 2.1 Score of 80.5 with Only 1.4M Rollouts, 70 Times Fewer than Reflection Beam

26 Oct
  1. MarkTechPost
    Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model

    Nanbeige (Nanbeige) released the 33B parameter dense large language model Nanbeige4.3-33B, supporting a context window of 524,288 tokens, leading in 17 out of 20 benchmark tests, including scoring 80.9 on Terminal-Bench 2.1, but lagging behind Qwen3.8-27B in SWE-Bench Pro.