Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding

Category narrowing to this degree really has diluted the term "SOTA" quite a lot

You mean it's state of the art for this subset of benchmarks you're targeting? No? Well surely it's state of the art for models of a certain size category? Huh, what about models that are available for local inference? O, state of the art for open models of a certain size made in the US specifically. Uh. Neat!

It's great to see another competitive open weight model. Congratulations!

But as is, it's probably hard to become popular amongst hobbyists and local model enthusiasts like Qwen's 27B class models or MOE models because it's a 65B dense model and inference will be prohibitively slow on consumer grade hardware, and seems like the goal/mission of the company behind this model is to automate chip design.

So if I have to assume and connect the dots, this is for the companies who want to use AI in air-gapped environments for chip design so that they protect their IPs (not leaking to third party AI labs/hostile parties) and can afford lots of server grade inference hardware for a quick unlock with the big dense model.

Neat!

Congratulations I guess but it's a pretty bad look to not even mention those other open weight models.

The hosted GMI cloud model on openrouter does not invoke confidence. Ignoring 131k context length limit the tool call failure rate is not acceptable.

When asked to call 'submit_html' tool for our bench it wrote the whole page as plain reply text, then called submit_html with "This is the complete submit trigger." 4/8 tools failed.

There are several situations where some of us are explicitly prohibited from using models from companies not under the jurisdiction of the United States. It doesn't mean these models are competitive when looked at from a global view but it is also not an irrelevant differentiation.

This model scores 80.5 on Terminal Bench 2.1

For reference Qwen 3.8 Flash Next (which is bigger than this model but probably cheaper to run due to MoE) scores 84.3, and Qwen 3.8-27B scores 73.0.

I think this model could be competitive, depending on how much it was benchmaxxed compared to Qwen?

I trained a tiny LLM earlier this year, it's fits under 10mb and it's completely useless, can I call it SOTA for my neighborhood?

When a post doesn’t mention Qwen for comparison, you know the performance isn’t going to be great.

I understand the desire to look the best in some self defined category. But not showing any comparison against the actual leaders in open weight renders all of the benchmarking completely information-less for the actual end users (who care about real performance comparison against the real field).

New model is always a good news!

However, the chart comparing the score-by-parameters with existing models is misleading, because the others are all MoE, and this model is not. The article describes it as an architectural decision to optimize against NVIDIA cards(which sounds reasonable, as US labs are relatively easy to gain access to these high-performance chips) but the chart should've been adjusted to reflect the difference, say, by using the heuristic that `effective param size ≈ sqrt(total param size * activiated param size` on MoEs.

Agentic coding benchmarks look great.

I think we will find the market is going to move into omni multi-modal models: text, moe, agentic tasks, voice, images, and video all in one LLM.

Wasn't ricursive's sthick to automate EDA and chip design with LLMs? Why are they releasing open weight agentic coding models and benchmarking on everything but EDA and chip design?

Speaking as someone who has High-Memory Mac Studios, it is great that there is finally a model that will run at usable speeds with high quality. In my experience, the sweet spot for Mac Clusters is around 200GB of weights, which this is far under. Congratulations to the recursive team for building an open model that will actually run!

Nice to see another open weight model.

I cannot evaluate how valuable this model is based on the performance table alone because the comparison is kind of narrow. Waiting for third-party benchmarks to provide more information.

I do not have the hardware to run it properly without super-heavy quantization (about q2). Besides, the architecture is not optimized for local usage (sliding window attention also requires KV cache). I think I will just move on.

Looks like this may be the GGUF one: https://huggingface.co/AtomicChat/Ripple-65B-Preview-GGUF

Oh, I thought this was from the company recursive.com who released the very good for its size model Flint.

I am looking at the HF model page https://huggingface.co/ricursiveAI/Ripple-65B-Preview and it looks like I need to wait a day for an Apple Silicon version.

So we have a Recursive and a Ricursive, both making LLMs?

These are pretty odd benchmarks to use. Terminal Bench 2.1 is old. SWE-bench was disavowed by OpenAI. DeepSWE is clearly benchmark-hacked by Gemini.

Oh yeah well I trained a bloom filter on this agentic coding benchmark and it selects the right answer with 99% accuracy

Looking forward to a "SOTA model from my basement". Make the category narrow enough, anybody can claim SOTA.

Global open-weight SOTA ... https://huggingface.co/moonshotai/Kimi-K3.1-Pro

Not comparing to Qwen3.8 27B gives cherry-picked benchmark comparison energy. Still, Ripple-65B outscores Qwen3.8 27B, according to the reported benchmarks.

Would love a 27B model. Wonder if we'll get a distillation.

Not to be confused with https://recursive.com/

Showing a terminal bench 2.1 score at the top is very misleading when terminalbench 4.0 is the current metric as a very quick glance for anything we care about for 30B to 180B size possible to locally-host models.

The comparison chart is befuddling.

Ouh this is the size of model I want to see more of (just got a 98gb m5u studio)

Ouh that fits right on a base M5U Studio! Excited to try it.

So the model is "meh" at best, but the cost of the training is interesting. So we can extrapolate that in 3 years the cost will be below $5m for a model like this.

It's very well within the capabilities of any reasonable organization, maybe even crowd-funded.

What is this thing where model comparisons blatantly pick other crap models and show themselves at the top left (cheap and effective)? Put all the open models, or put all the cheap models, or whatever, but just cherry picking is egregious.

[dead]

[dead]

[flagged]

I'm sorry but what is "U.S." implying here?

Is this sponsored by the us gov? did it only train on US data? does it only work on the US?