引入 ripple- 65B 预览
边框开放重量模型
Ricursive Intelligence是一款通过协同优化硬件，软件和模型来建立自改进系统的全装人工智能实验室。
为了给芯片设计平台提供动力，我们从零开始训练一种语言模型，在代理编码上设置了一个新的美国开放量的SOTA. 我们优化了我们的培训和服务平台的模型架构，推动大幅度提高效率。
今天，我们发布了Ripple-65B预览，这是专有模型的通用变体，它为我们的AI芯片设计平台提供了动力。 该模型在12兆令牌上预训，支持的上下文长度可达1M. Ripple-65B预览在代理编码上产生强烈结果，在Bench 2.1终端上实现了美国开放重量的最先进(80.52)，并超过了反射Beam 501B(80.1),Meta Muse Spark Open 816B(76.4)，思维机Inkling 975B(63.8)和NVIDIA Nemotron 3 Ultra 550B(56.4)等显著较大的美国开放重量模型。
Ripple的密集模型架构提供了强大的推理性能，低纬度符号生成。 该模型的小型参数信封也使得它能够在NVIDIA DGX Spark等本地设备上高效运行，这些设备具有相对于数据中心加速器的高内存带宽低的高计算FLOP.
Ripple-65B预览可在Hugging Face,Together AI,GMI云和OpenRouter上获得。 烟火今天晚点开始直播
以第10种成本实现美国开放-重量前沿
LLM预训传奇性昂贵。 2023年，OpenAICEO萨姆·阿尔特曼·称认为GPT-4的训练费用超过1亿美元。 次年，边疆AI型号的Dario Amodei 称可以花费高达10亿美元进行训练，预计单是前期训练的费用到2026年至2027年将增加到10-100B.
相比之下，我们只以32M美元开发并训练了Riple-65B预览，这是我们所了解的关于代理编码的前一次开放量SOTA成本的十分之一。 注意32M是我们从头开始的全部研发成本，包括预训阶梯开发，数据混合，evals，调试，合成数据生成，后培训食谱开发，以及最终的英雄预训和后培训运行。
深化强化学习
而不是投资一个野蛮的RL爬升，我们仔细地重温了我们的SFT和RL配方，重点是计算效率。 我们选择不购买任何培训后数据，而是设计自己的合成数据管道、质量过滤和覆盖度指标。
我们的混合RL配方采用短重采样，分数中心，线性长度判罚与高质量的长视合成环境。 RL任务适应我们的模型能力，随着模型的改进，我们不断增加更难的任务。
最终结果是RL从一个强大的最初的SFT检查站陡然攀升。 我们达到了80.5分的终极奔驰2.1分，仅推出1.4M，大约比之前的SOTA的反射Beam少70倍。
共同指定模型架构
Ripple-65B Preview的架构被设计为在NVIDIA硬件上接受训练并高效服务。 一项关键目标是作出设计决定，以减少培训前和培训后基础设施的复杂性，同时仍能产生强有力的推理和代理性表现。 因此，我们选择了训练一种现代化的密集变形器模型，而不是像其他实验室一样使用Mixture-of-Experts(MOE). 这一决定很快得到回报，我们从零开始训练一个强大的模型，同时在整个预训和后训练中保持了一个坚实的MFU(40%)，即使有简单的硬化，没有定制内核，也没有量化。
Rip65B 预览采用薄而深的模型宽度比，窄的d model(5120)和共80层。 为了提高推理性能，我们过多地提供关注负责人的数量，同时增加西西南-全球层比，并减少西西南的窗口规模，以补偿增加的费用。 此外，我们增加了FFN激活剪辑，关卡注意，全球NoPE，以及价值残存。
谨慎的模型疏导到大改进
我们通过一系列仔细的模型梯级实验，优化了Ripple的预训配方。
我们的起点是奥尔莫3号的模型梯子，我们从中测试了小规模的干预措施，重点是提高编码能力。 我们的v1.0梯级实施了StepFun批量大小和学习速率缩放定律，v1.1则实施了我们的新梯级，平稳地测量建筑尺寸和模型深度。 梯子v1.2从后规范基线切换为前+后规范，在小尺度上是中性的，但从4B开始显示出改进。 梯子v1.3 已执行 Sigmoid门注意。 Ladder v1.4 (2-stream mHC)在较小的尺度上表现出良好的效果，但在4B尺度上没有改进，因此被拒绝。 梯子 v1.5 切换到全局NoPE，并增加了最终LR冷却到零，梯子 v1.6 增值残基。 在最后版本中，我们切换到自定义的预训数据集，并附有大量改进的代码。
复合变化加在一起，提高了Dolma stack CE在4B模型尺度上测量的代码计算效率，最高可达16x. 换句话说，我们最后的Riple模型梯级可以用来实现与奥尔莫3号梯级相似的代码性能，而16x的计算较少。 注意奥尔莫3号梯级基线的坡度比梯级v1.0+更浅，显示表示计算效率的提高会随着额外的尺度而增加。
结论
我们建造了Ripple-65B预览，开始关闭模型和燃料硬件之间的环路。 这个模型的一个内部变体给我们的AI芯片设计平台带来力量我们期待着看到我们向前沿发展时出现的能力
我们有幸在现有的开源贡献的基础上更上一层楼，包括来自艾伦学院AI(AI2)的奥尔莫团队，Prime Intellect, RadixArk，和NVIDIA. 一个强大的开源生态系统也在自我改善，其贡献者加强生态系统，而生态系统又吸引了更多的贡献者。
在Ricursive，我们正在有选择地雇用杰出的研究人员和工程师，因为我们把完全的AI的前沿从模型推向硅。 如果您有兴趣，请查看https://www.ricursive.com/careers。
1名美国开放量级球员如Reflection(英语：Reflectionion)被报道承诺每月150M美元在SpaceX Colossus进行计算以训练他们的型号。 反省报道说，Beam需要4周的预训时间和4周的预训时间。
(原始内容存档于2026-09-26). Ricursive Intelligence, Inc.
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Introducing Ripple-65B Preview,
a Frontier Open-Weight Model
Ricursive Intelligence is a full-stack artificial intelligence lab building self-improving systems by co-optimizing hardware, software, and model.
To power our chip design platform, we trained a language model from scratch that sets a new U.S. open-weight SOTA on agentic coding. We optimized the model architecture for our training and serving platforms, driving substantial efficiency gains.
Today, we release Ripple-65B Preview, a general-purpose variant of the proprietary model that powers our AI chip design platform. The model is pretrained on 12 trillion tokens with support for up to 1M context length. Ripple-65B Preview delivers strong results on agentic coding, achieving U.S. open-weight state-of-the-art (80.52) on TerminalBench 2.1 and surpassing significantly larger US open-weight models such as Reflection Beam 501B (80.1), Meta Muse Spark Open 816B (76.4), Thinking Machines Inkling 975B (63.8), and NVIDIA Nemotron 3 Ultra 550B (56.4).
Ripple’s dense model architecture provides strong reasoning performance with low-latency token generation. The model’s small parameter envelope also enables it to run efficiently on local devices such as NVIDIA DGX Spark, which feature high compute FLOPs with low memory bandwidth relative to datacenter accelerators.
Ripple-65B Preview is available on Hugging Face, Together AI, GMI Cloud, and OpenRouter. Fireworks goes live later today.
Achieving U.S. Open-Weight Frontier at One-Tenth Cost
LLM pretraining is legendarily expensive. In 2023, OpenAI CEO Sam Altman said that GPT-4 training costs exceeded $100 million. In the following year, Dario Amodei said that frontier AI models can cost up to $1 billion to train, projecting that costs would rise to $10-100B by 2026 to 2027 for pretraining alone.
In contrast, we developed and trained Ripple-65B Preview for only $32M, which is one-tenth the cost of the previous open-weight SOTA on agentic coding, as estimated to the best of our knowledge.¹ Note that $32M is our full R&D cost starting from scratch, including pretraining ladder development, data mixing, evals, debugging, synthetic data generation, posttraining recipe development, and the final hero pretraining and posttraining run.
Steep Ascent Reinforcement Learning
Rather than investing in a brute-force RL climb, we carefully iterated on our SFT and RL recipes with a focus on compute efficiency. We opted not to buy any posttraining data, instead devising our own synthetic data pipelines, quality filtering, and coverage metrics.
Our mixed RL recipe uses truncated importance sampling, score centering, and linear length penalty with high-quality long-horizon synthetic environments. The RL tasks are tuned for our model capabilities, and we continuously added harder tasks as the model improved.
The end result was a steep RL climb from a strong initial SFT checkpoint. We reached a TerminalBench 2.1 score of 80.5 with only 1.4M rollouts, roughly 70x fewer rollouts than Reflection Beam, the prior SOTA.
A Co-Designed Model Architecture
Ripple-65B Preview’s architecture was designed to be trained and served efficiently on NVIDIA hardware. A key objective was to make design decisions which reduce pretraining and posttraining infrastructure complexity while still enabling strong reasoning and agentic performance. Therefore, we chose to train a modernized dense Transformer model rather than use Mixture-of-Experts (MoE) like other labs. This decision quickly paid off, and we were able to train a strong model from scratch while maintaining a solid MFU (40%) throughout pretraining and posttraining, even with simple sharding, no custom kernels, and no quantization.
Ripple-65B Preview uses a thin-and-deep model aspect ratio, with a narrow d_model (5120) and 80 total layers. To improve reasoning performance, we overprovision the number of attention heads while increasing the SWA-to-global layer ratio and reducing SWA window size to compensate for the increased cost. In addition, we added FFN activation clipping, gated attention, global NoPE, and value residuals.
Careful Model Laddering Led to Large Improvements
We optimized Ripple’s pretraining recipe through a series of careful model laddering experiments.
Our starting point was Olmo 3’s model ladder, from which we tested interventions at small scale with a focus on improving coding capabilities. Our v1.0 ladder implemented StepFun batch size and learning rate scaling laws, and v1.1 implemented our new ladder which smoothly scales architectural dimensions and model depth. Ladder v1.2 switched from the post-norm baseline to pre+post norm, which was neutral at small scale but showed improvements starting at 4B. Ladder v1.3 implemented sigmoid gated attention. Ladder v1.4 (2-stream mHC) showed good results at smaller scale but did not improve at 4B scale, and was therefore rejected. Ladder v1.5 switched to global NoPE and added a final LR cooldown to zero, and ladder v1.6 added value residuals. In the final version, we switched to our custom pretraining dataset with substantially upsampled code.
Together, the compounded changes improved compute efficiency on code as measured by dolma_stack CE by up to 16x at 4B model scale. In other words, our final Ripple model ladder can be used to achieve similar code performance to the Olmo 3 ladder with 16x less compute. Note that the slope of the Olmo 3 ladder baseline is shallower than ladder v1.0+, which shows that the compute efficiency improvement will increase with additional scale.
Conclusion
We built Ripple-65B Preview to begin closing the loop between models and the hardware that fuels them. An internal variant of this model powers our AI chip design platform, and we look forward to seeing the capabilities that emerge as we scale toward the frontier.
We were fortunate to build on top of existing open-source contributions, including those from the Olmo team at the Allen Institute for AI (AI2), Prime Intellect, RadixArk, and NVIDIA. A strong open-source ecosystem is also self-improving, where contributors strengthen the ecosystem, which in turn draws more contributors.
At Ricursive, we are selectively hiring exceptional researchers and engineers as we push the frontier of full-stack AI from model to silicon. If you’re interested, check out https://www.ricursive.com/careers.
¹ U.S. open-weight players like Reflection were reported to commit to $150M per month for compute at SpaceX Colossus to train their models. Reflection reported that Beam took 4 weeks to pretrain and 4 weeks to posttrain.
© 2026 Ricursive Intelligence, Inc.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.