| ▲ | Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding(ricursive.com) |
| 178 points by Philpax 4 hours ago | 57 comments |
| |
|
| ▲ | advael 3 hours ago | parent | next [-] |
| 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! |
| |
| ▲ | gpm 2 hours ago | parent | next [-] | | Playing games like this also makes it very easy to accidentally say things that aren't true, like in the headline of the article... > Introducing Ripple-65B Preview, a Frontier Open-Weight Model It's not "frontier", it's not "frontier" even in the category of open-weight models (though usually that phrasing would mean frontier over all, and also open weight)... it's only "frontier" in the small sub-category of US made open-weight models. | | |
| ▲ | GuB-42 an hour ago | parent [-] | | It is technically frontier, if by "frontier" we mean the Pareto frontier. Being at the Pareto frontier means that you can't get something better on one aspect without compromising on some other aspect. The trick is that it is easy to be at the Pareto frontier by having enough dimensions. Here, we have 4: score, number of parameters, origin (US) and open-weight. You can get a better score or a lower number of parameters, but you have to compromise on at least one of the other aspects. That's a pretty weak frontier if you ask me. | | |
| ▲ | derefr an hour ago | parent [-] | | “Origin” isn’t part of the Pareto frontier because it’s not a quantitative dimension. (You could say that “ability to be trusted by US government/military users” is quantifiable, but I would argue that by that metric, this model falls behind non-US-originated open-weight models that have more eyes on them!) |
|
| |
| ▲ | quixoticelixer- 40 minutes ago | parent | prev | next [-] | | Fourth most popular comedy folk rock duo | |
| ▲ | yorwba 2 hours ago | parent | prev | next [-] | | AFAIK the currently biggest US open-weight model is Inkling, which they include in the comparison, so at least their claim is not limited to a size category. Still they score 0.42 percentage points higher than the second-best model on their chosen benchmark, so it's hardly a meaningful lead. | |
| ▲ | visarga 3 hours ago | parent | prev | next [-] | | No, the term is well defined. It means being on the Pareto curve of performance vs size, such as: https://tinyurl.com/26sw35fw | | |
| ▲ | advael 2 hours ago | parent | next [-] | | Irrelevant. A term can be well-defined and used correctly while also being diluted. In this case, I am pointing out that it's diluted by narrowing the category in which the claim is being made. I have no doubt that in this narrowed category, the term is being used correctly. This has no bearing on the fact that by narrowing the category in that way, this claim is made remarkably less impressive or interesting to many people hearing it, including myself | |
| ▲ | O_H_E 3 hours ago | parent | prev | next [-] | | well yes but the y axis heavily depends on the benchmark you choose | |
| ▲ | epolanski 2 hours ago | parent | prev [-] | | By...whom? An ISO standard? Highly cited scientific literature? A tinyurl pic linked by randoms over the internet? |
| |
| ▲ | frinxor 2 hours ago | parent | prev | next [-] | | The US bit got me as well. I guess it makes sense - there is a segmentation where users may want to run on domestic "frontier LLMs." | |
| ▲ | swiftcoder 2 hours ago | parent | prev | next [-] | | For sure, but unfortunately, there's a non-zero chance the state department at some point makes the "US" part of this mandatory for a bunch of companies. | |
| ▲ | dev_l1x_be 2 hours ago | parent | prev [-] | | State of the AI-Apocalypse. |
|
|
| ▲ | stewartpark 3 hours ago | parent | prev | next [-] |
| 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! |
| |
| ▲ | 0x00cl 19 minutes ago | parent | next [-] | | Maybe is a bit more niche, but I'm pretty sure there is plenty of people who have enough RAM to run dense models like this one, I don't know how slow is slow but people are willing to sacrifice speed if it means they get a good model running locally. Look at "SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF" on HuggingFace, It says it has 3.5M downloads this month and its offering a model that is 2-bit and 3-bit quantized (66GB - 83.9GB). If this gets quantized to 4-bit (Q4_0) which usually there is very little loss in intelligence compared to the base model, then the model will probably be around 40GB, and looking that people are downloading a model that is at least 20GB bigger, then it might allow even more people to run a bigger model. But it will only be popular among hobbyist if its as good as it claims to be. | |
| ▲ | 59nadir 2 hours ago | parent | prev | next [-] | | Is it competitive overall or just competitive when compared to "U.S. Open-Weight models that are otherwise pretty decent at agentic coding"? The latter does not imply the former. | |
| ▲ | regularfry 2 hours ago | parent | prev [-] | | What'll be interesting is how it performs relative to Qwen3.8 27b when quantised to the same size. Which, of course, is a comparison they don't want to do. | | |
| ▲ | 40four 2 hours ago | parent [-] | | Well, obviously. These newer American models aren’t competing with Qwen yet. They are competing for American businesses and organizations that can’t/won’t use Chinese models. They are playing catch-up, but I’m happy to see new players to the game, and it will be interesting to watch how they develop |
|
|
|
| ▲ | wmf 3 hours ago | parent | prev | next [-] |
| Congratulations I guess but it's a pretty bad look to not even mention those other open weight models. |
| |
| ▲ | karmakaze 3 hours ago | parent | next [-] | | What's funnier is that after cherry picking models for comparison, they further cherry picked Inking for size comparison instead of better and smaller Beam 1.0 The bar chart also shows Nemotron 3 Ultra instead of smaller and better Nemotron 3.5 Super. | |
| ▲ | tedivm 3 hours ago | parent | prev | next [-] | | We've always been at war with eurasia. Seriously though, it's just easier to pretend that China doesn't exist when they're stomping us this badly. | |
| ▲ | stewartpark 2 hours ago | parent | prev | next [-] | | I think it's because the primary users of this model won't even look at "those other open weight models" as options. This is the right chart to present to those who would consider this model. (e.g. US Gov't agencies, chip design industry, etc etc) | |
| ▲ | petcat 2 hours ago | parent | prev | next [-] | | Do those other models usually mention all the competition? I haven't noticed | |
| ▲ | schleck8 2 hours ago | parent | prev | next [-] | | Because Chinese models heavily rely on distillation and regularly behave strangely in public benchmarks (where they score much better than in private evaluations), even if many world class researchers are Chinese. It's difficult to quantify to what extend labs like Moonshot would be competitive if they could not distill from OpenAI, now that Anthropic mostly blocks it https://www.theinformation.com/articles/openais-astra-model-... | |
| ▲ | flatline 3 hours ago | parent | prev [-] | | Right? These are not likely to be meaningful points of comparison to anyone but the authors. | | |
| ▲ | wmf 3 hours ago | parent [-] | | Maybe Palantir will use it or something. There probably is a market for US models but I don't know how large it will be. | | |
| ▲ | Onawa 3 hours ago | parent [-] | | US Gov won't let Chinese models be hosted on any internal infra, not even supposed to use them for local AI. So there is absolutely a market for US-made or non-Chinese models there. |
|
|
|
|
| ▲ | bluejay2387 2 hours ago | parent | prev | next [-] |
| 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. |
| |
|
| ▲ | 2001zhaozhao 3 hours ago | parent | prev | next [-] |
| 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? |
|
| ▲ | antoineMoPa an hour ago | parent | prev | next [-] |
| 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? |
|
| ▲ | thesnarkitecht 2 hours ago | parent | prev | next [-] |
| 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! |
|
| ▲ | minimaxa 2 hours ago | parent | prev | next [-] |
| 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. |
|
| ▲ | syntaxing 3 hours ago | parent | prev | next [-] |
| When a post doesn’t mention Qwen for comparison, you know the performance isn’t going to be great. |
| |
|
| ▲ | a11r an hour ago | parent | prev | next [-] |
| 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). |
|
| ▲ | purplemantis 2 hours ago | parent | prev | next [-] |
| 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. |
|
| ▲ | mrinterweb an hour ago | parent | prev | next [-] |
| Not comparing to Qwen3.8 27B gives cherry-picked benchmark comparison energy. Still, Ripple-65B outscores Qwen3.8 27B, according to the reported benchmarks. |
|
| ▲ | dofm 3 hours ago | parent | prev | next [-] |
| So we have a Recursive and a Ricursive, both making LLMs? |
| |
| ▲ | wmf 3 hours ago | parent | next [-] | | Time to start Rickursive, the lab that will never give you up and never let you down. | | | |
| ▲ | giwook 3 hours ago | parent | prev | next [-] | | There seems to be some recursion going on here. | |
| ▲ | xyzrt 3 hours ago | parent | prev [-] | | One making self-improving AI via AI-chip co-design, the other relying only on software for self-improving AI. | | |
| ▲ | dofm 3 hours ago | parent [-] | | I hope they never have to call each other about anything. It could get confusing if neither knows who is the base case. |
|
|
|
| ▲ | jszymborski 3 hours ago | parent | prev | next [-] |
| Would love a 27B model. Wonder if we'll get a distillation. |
|
| ▲ | RGS1811 3 hours ago | parent | prev | next [-] |
| The comparison chart is befuddling. |
|
| ▲ | soundworlds 3 hours ago | parent | prev | next [-] |
| Not to be confused with https://recursive.com/ |
|
| ▲ | cyberax an hour ago | parent | prev | next [-] |
| 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. |
|
| ▲ | 6thbit 3 hours ago | parent | prev [-] |
| 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? |
| |
