Update README.md

README.md

@@ -11,13 +11,103 @@ Ripple-65B was codesigned to be trained and served efficiently on NVIDIA hardwar
 For our first model, a key design objective was to make model architecture 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).
-This decision quickly paid off, and we were able to train a strong model from scratch with a small team with only 5 people.
+This decision quickly paid off, and we were able to train a strong model from scratch with a small team with only 5 people 
+while maintaining solid MFU throughout pretraining and posttraining.
 
 Starting from the Olmo llama-like baseline, we switch to a more Gemma-like architecture to improve training stability at scale. 
 Ripple-65B 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 ratio and reducing SWA window size to compensate.
+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.
 In addition, we added FFN activation clipping, gated attention, global NoPE, and value residuals.
 
 The end result is a model that is especially capable at reasoning given its model size.
 At 65B total parameters, the model can fit on consumer hardware, such as the NVIDIA DGX Spark.
-The model is also fast at datacenter token generation, easily reaching 250+ tok/s with DFlash2 on 4xB200. 
+The model is also fast at token generation when running on datacenter GPUs, easily reaching 250+ tok/s with DFlash2 on 4xB200 prior to quantization. 
+
+**Pre-training.** Ripple-65B Preview is pretrained from scratch on a 12T text-only corpus at 8K sequence length.
+
+**Post-training.** We use a standard post-training recipe consisting of midtraining, SFT, and mixed domain RL. 
+
+## Instruct Model
+
+
+<div align="center">
+
+| Benchmark (Metric) | Opus-5.0 | GPT-5.6 Sol | K3 | GLM-5.3 | DS-V4-Pro | DS-V4-Flash | DS-V4.1-Flash |
+| :--- | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
+| **Reasoning** | | | | | | | |
+| GPQA Diamond (Pass@1) | 93.4 | **94.1** | 92.9 | 88.1 | 92.4 | 89.9 | 90.9 |
+| HLE (Pass@1) | **56.3** | 44.5 | 43.5 | 42.0† | 42.7† | 37.8† | 36.8 (39.1†) |
+| Codeforces (Rating) | — | — | — | — | 3348 | 3289 | **3471** |
+| MathArena Apex (Pass@1) | — | — | **65.6** | — | 65.3 | 58.6 | **65.6** |
+| **Agentic** | | | | | | | |
+| Terminal-Bench 2.1 (Pass@1) | 89.1 | 88.8 | 88.3 | 88.2 | 87.9 | 82.7 | **90.6** |
+| Terminal-Bench 3.0 (Pass@1) | **43.3** | 34.4 | 17.7 | 28.3 | 11.8 | 7.6 | 30.0 |
+| Terminal-Bench 4.0 (Pass@1) | **51.8** | 39.9 | 12.6 | 37.9 | 12.4 | 7.0 | 31.2 |
+| DeepSWE v1.1 (Resolved) | 74.0 | 73.0 | 67.5 | 66.9 | 62.7 | 54.4 | **74.2** |
+| ProgramBench (Almost@1) | **37.0** | 23.0 | 17.5 | 19.0 | 15.5 | — | 20.3 |
+| NL2Repo-Bench (Score) | **75.3** | 56.8 | 58.0 | 58.0 | 61.5 | 54.2 | 64.0 |
+| CyberGym (Pass@1) | — | 84.5 | 80.0 | 84.5 | 83.3 | 76.7 | **88.1** |
+| SEC-Bench Pro (Pass@1) | — | **74.3** | — | — | 56.4 | 30.9 | 62.8 |
+| ExploitGym (Pass@1) | 22.1 | **33.7** | — | 15.0 | 5.4 | 1.8 | 15.3 |
+| HLE w/ tools (Pass@1) | 63.6 | — | 59.8 | 62.5 | 60.0 | 51.5 | **63.9** |
+| AutomationBench (Pass@1) | 50.3 | 45.8 | 46.7 | 48.8 | 43.2 | 37.7 | **54.8** |
+| Agent's Last Exam (Pass@1) | 28.6 | 26.7 | 27.6 | 28.5 | 25.7 | 25.2 | **31.8** |
+| Chartography w/ tools (Pass@1) | **84.0** | 79.9 | 68.1 | — | — | — | 78.9 |
+| BabyVision w/ tools (Pass@1) | **94.1** | 88.9 | 85.7 | — | — | — | 89.6 |
+| ZeroBench-main w/ tools (Pass@5) | 52.0 | **53.0** | 41.0 | — | — | — | 49.0 |
+
+</div>
+
+
+## Base Model
+
+We use an internal fork of [Olmes](https://github.com/allenai/olmes) for our evals.
+
+<div align="center">
+
+| Benchmark (Metric) | # Shots | DeepSeek-V4-Flash-Base | DeepSeek-V4-Pro-Base | DeepSeek-V4.1-Flash-Base |
+| :--- | :---: | :---: | :---: | :---: |
+| Architecture | — | MoE | MoE | MoE |
+| # Backbone Params | — | 284B | 1.6T | 552B |
+| # Activated Params | — | 13B | 49B | 8B / 16B |
+| **World Knowledge** | | | | |
+| AGIEval (EM) | 3–5-shot | 83.9 | **84.4** | 83.4 |
+| MMLU-Pro (EM) | 5-shot | 68.3 | 73.5 | **74.1** |
+| C-Eval (EM) | 5-shot | 92.1 | **93.1** | 92.1 |
+| MultiLoKo (LLM-Judge) | 5-shot | 42.6 | **50.9** | 45.5 |
+| SimpleQA-Verified (EM) | 25-shot | 30.1 | **55.2** | 42.3 |
+| SuperGPQA (EM) | 5-shot | 46.5 | **53.9** | 53.1 |
+| **Language & Reasoning** | | | | |
+| BBH (EM) | 3-shot | 86.9 | **87.5** | 86.1 |
+| BBEH (EM) | 1-shot | 25.4 | **29.8** | 27.2 |
+| DROP (F1) | 1-shot | **88.6** | **88.7** | 87.9 |
+| HellaSwag (EM) | 0-shot | 85.7 | **88.0** | 87.2 |
+| **Code & Math** | | | | |
+| BigCodeBench (Pass@1) | 3-shot | 56.8 | 59.2 | **60.6** |
+| HumanEval (Pass@1) | 0-shot | 69.5 | 76.8 | **79.4** |
+| GSM8K (EM) | 8-shot | 90.8 | 92.6 | **93.0** |
+| MATH (EM) | 4-shot | 57.4 | **64.5** | 61.1 |
+| MGSM (EM) | 8-shot | **85.7** | 84.4 | 80.2 |
+| **Long Context** | | | | |
+| LongBench-V2 (EM) | 1-shot | 44.7 | **51.5** | 45.2 |
+| **Multimodal** | | | | |
+| MMMU-Pro (EM) | 4-shot | — | — | 56.5 |
+| CVBench (EM) | 4-shot | — | — | 77.9 |
+| DocVQA (LLM-Judge) | 4-shot | — | — | 95.6 |
+| RefCOCO-avg (Acc@0.5) | 0-shot | — | — | 86.0 |
+
+</div>
+
+## Citation
+
+```bibtex
+@misc{ricursive2026ripple65b,
+      title={Ripple-65B},
+      author={Ricursive Intelligence},
+      year={2026},
+}
+```
+
+## Contact
+
+For questions and comments, please contact us at [info@ricursive.com](mailto:info@ricursive.com). To learn more, please visit our [website](https://www.ricursive.com).