whatmodel.sh / Ripple-65B Preview
Ripple-65B Preview
↗ share
+ contribute
Ripple-65B Preview is an open-weights language model from Ricursive Intelligence, released November 2026. Its published size is 65B parameters and a 1M-token context window. Licence: apache-2.0.
Catalogue identity notes
BF16 instruction-tuned checkpoint (this repo, created 2026-11-04, 129.6 GB). Ricursive also publishes Ripple-65B-Preview-NVFP4 ('Experimental'; 'NVFP4 for the MLP layers while keeping attention in BF16'; 57.2 GB; its card gives no serving command), the pretrain-only Ripple-65B-Base-Preview (not added) and a DFlash2 draft. Text-only. Card: 'Our Preview release has gone through relatively light posttraining, with a focus on agentic coding.' Launch blog (ricursive.com/ripple): 'available on Hugging Face, Together AI, GMI Cloud, OpenRouter, and Fireworks' - no first-party API and the host pages were not checked, so api is null. The BF16 repo's card has no vLLM/SGLang command (only custom inference/server code), so local routes are the official Ollama library tags.
Model information sources
Release date ↗ · Weights licence ↗ · Parameter count ↗
Dense. Card table: Total params 65B, Active params 65B. HF safetensors metadata at b172323: 64,782,689,360 BF16 parameters.
Score within the LMArena (text) comparison set only.
$ how it compares
The LLMs ranked either side of Ripple-65B Preview, out of 1206 with a score.
| # | model | score | where |
|---|---|---|---|
| 848 | 44±2 | local | |
| 849 | 44±1 | local | |
| 850 | Ripple-65B Preview | 44±7 | local |
| 851 | 44±1 | local | |
| 852 | 44±1 | local |
$ what it's good at
agent tasksnot ranked on this task yet · 6 tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| SWE-bench Verified lab | 76.5% ✎ | ▁▁▁▁▁▁▂▄█▆▁▁ beats 114/140 | — |
| Terminal-Bench 2.1 lab | 80.5% ✎ | ▁·▁·▁▁▄▃▅█▆█ beats 46/68 | — |
| SWE-bench Pro (split unspecified) lab | 61.4% ✎ | ▁·▁▁▁▁▁▄█▁▁▁ beats 48/64 | — |
| MCP-Atlas (public set) lab | 71.7% ✎ | ▂·▁▂▂·▁▃█▆▆▃ beats 27/41 | — |
| DeepSWE v1.1 (resolved) lab | 44.3% ✎ | ▂····▃▁▁▄█▆█ beats 4/36 | — |
| τ³-bench Banking lab | 31.3% ✎ | ▂▂·▅█▂▃▃▃▆▂▆ beats 23/25 | — |
codingnot ranked on this task yet · 4 tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| SWE-bench Verified lab | 76.5% ✎ | ▁▁▁▁▁▁▂▄█▆▁▁ beats 114/140 | — |
| SWE-bench Pro (split unspecified) lab | 61.4% ✎ | ▁·▁▁▁▁▁▄█▁▁▁ beats 48/64 | — |
| SciCode (level unspecified) lab | 39.4% ✎ | ▁····▃▃▄▄█▆▃ beats 27/42 | — |
| DeepSWE v1.1 (resolved) lab | 44.3% ✎ | ▂····▃▁▁▄█▆█ beats 4/36 | — |
mathnot ranked on this task yet · 4 tests
other tests6 not in a task · 77% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Icelandic LLM Leaderboard average (Miðeind, Sept 2026 suite) | 7.7% ✎ | ▃▂▄▂▃▃▂▇▇█▇▁ beats 1/47 | 77% |
| GPQA Diamond lab | 85.2% ✎ | ▁▁▁▁▂▃▅▆▇▇█▂ beats 235/336 | — |
| MMLU lab | 89.6% ✎ | ▁▁▁▁▁▂▂▅██▆▄ beats 205/215 | — |
| MMLU-Pro lab | 84.2% ✎ | ▁▁·▁▁▁▃▂▄▅█▁ beats 163/189 | — |
| IFBench (aggregation unspecified) lab | 79.2% ✎ | ▂▆▂▃▄▄▂▃▄▆█▇ beats 79/88 | — |
| AutomationBench (Public) lab | 36.4% ✎ | █····▄··▄·▄▄ beats 3/5 | — |
$ where to run it
Memory is what a version takes in GPU memory (or a Mac's unified memory) with an 8k-token conversation. With less, it runs partly from system RAM, more slowly. Enter your machine to see what runs at full speed.
Ollama · recommended
version
- Install Ollama
download it (macOS / Windows) ↗
Linuxcurl -fsSL https://ollama.com/install.sh | sh - Run Ripple-65B Preview · needs 42.2 GB memory, 38.8 GB download · from ollama.comchat in the terminal
ollama run ripple-preview:65b-q4_K_M
Model repository and documentation: huggingface.co/ricursiveAI ↗
$ how the score was worked out · Miðeind Icelandic LLM Leaderb… 77%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
| 1 | 77% | |
| prior · pull toward the middle | 23% | |
| weighted average | = 44.0 |
not counted (16)
| ricursiveAI/Ripple-65B-Preview model card | MMLU [correct] | 89.6% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | MMLU-Pro [correct] | 84.2% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | GPQA Diamond [correct] | 85.2% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | AIME 2024 [correct] | 97.1% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | AIME 2025 [correct] | 98.3% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | AIME 2026 [correct] | 95.0% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | HMMT February 2026 [correct] | 90.5% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | Terminal-Bench 2.1 [correct] | 80.5% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | SWE-bench Verified [correct] | 76.5% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | SWE-bench Pro (split unspecified) [correct] | 61.4% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | DeepSWE v1.1 (resolved) [correct] | 44.3% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | SciCode (level unspecified) [correct] | 39.4% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | τ³-bench Banking [correct] | 31.3% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | AutomationBench (Public) [correct] | 36.4% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | MCP-Atlas (public set) [correct] | 71.7% | fewer than three models have this test yet, or this source ran it for this model alone |
| ricursiveAI/Ripple-65B-Preview model card | IFBench (aggregation unspecified) [correct] | 79.2% | fewer than three models have this test yet, or this source ran it for this model alone |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json