| ▲ | DaisyRose-27B: 1,200-task RL took Qwen3.8 from 5.6% to 12% on Terminal-Bench 4.1(radixark.com) |
| 83 points by de_sty 3 hours ago | 18 comments |
| |
|
| ▲ | woah an hour ago | parent | next [-] |
| The heavy reliance on eval datasets showcased here and also evidenced by the frontier lab's high payments for such data cast doubt on the claims that AI will become self-improving soon |
|
| ▲ | xlayn 2 hours ago | parent | prev | next [-] |
| Snap, no gguf to download, will wait for one to appear to test, any improvement over an already good model is good news. "18 days on 144 H200 GPUs" insane.. didn't thought so much hardware would be needed to train Qwen. |
| |
| ▲ | drdaeman 37 minutes ago | parent [-] | | Rather than waiting, ask your assistant to quantize for you. It could be not the best quantization possible but it’s relatively quick, and I had an overall positive experience. | | |
| ▲ | smallerize 22 minutes ago | parent [-] | | It is already quantized. You need to convert to GGUF format though. Did your assistant explain this wrong? | | |
| ▲ | drdaeman 18 minutes ago | parent [-] | | My bad, I haven’t really looked yet (and haven’t asked any assistants), thought it’s BF16 safetensors as usual, and for those I used to quantize to ggufs with just a couple commands. |
|
|
|
|
| ▲ | jacquesm an hour ago | parent | prev | next [-] |
| I always wonder what you lose when you start maxing out some specific benchmark. It is really hard to do this in a balanced way and I would love if everybody that did these kind of efforts would take the time to also list the before/afters of a number of other benchmarks that they did not use in their training. |
| |
| ▲ | nblintao 30 minutes ago | parent | next [-] | | Good point! First of all, none of the benchmark tasks were used in training, though the training data was designed to improve capabilities on tasks like TB4.1. If you consider TB4.1 the main target, the blog actually shows before/after results on several other benchmarks, with no regressions and even some improvements. | | |
| ▲ | jacquesm 25 minutes ago | parent [-] | | Yes, I've seen that, but I wonder what is missing because I find it hard to believe that you can just do some RL and then end up with a model that is better across the board. I would find it far more believable if you outperformed on the benchmark that was your goal (assuming you selected training data to give you an edge there) and underperformed on things that you were not optimizing for. But since it all moves in one direction (better) I feel there must be some kind of unknown at play here. Even so, task specific AI is definitely the future, you don't hire a plumber to do your bookkeeping either... | | |
| ▲ | reissbaker 17 minutes ago | parent [-] | | Why do you find it hard to believe that you can do some RL and improve across the board (albeit less strongly on other semi-related tasks than your objective)? That's the idea behind generalization: it's not perfect but you get a little bit of cross-task generalization, so if you just keep spamming more RL environments you can hill-climb to AGI. That's what the labs are doing! | | |
| ▲ | jacquesm 10 minutes ago | parent [-] | | You can hill climb but you will (or at least, I think you will!) lose generality. And no, that won't get you to AGI, that is a completely different goal. |
|
|
| |
| ▲ | gardnr 34 minutes ago | parent | prev [-] | | Check the article. It’s the first graphic that has other benchmarks. | | |
|
|
| ▲ | robertkarl 2 hours ago | parent | prev | next [-] |
| I look forward to testing this against my private eval set to determine if it's benchmaxxing or actual improvement based on just 1200 tasks. downloading it now |
|
| ▲ | XCSme 2 hours ago | parent | prev | next [-] |
| Is it post-training or bench-maxxing? |
|
| ▲ | octoberfranklin 2 hours ago | parent | prev [-] |
| FWIW that is about $331,000 on low-onboarding-grief clouds (e.g. vast.ai) right now using the SXM variants they used. You might be able to shave that down using a different flavor of H200 (like PCIe) but it would take more hours, so benefits are unclear. |
| |
| ▲ | lumost an hour ago | parent [-] | | Savings would likely be larger with alternative accelerators such as trainium, no? | | |
| ▲ | octoberfranklin 38 minutes ago | parent [-] | | Savings would likely be larger with magic beans such as floobarbaz, no? But seriously. If there were a cheaper and reliable way, I'm sure the authors would have mentioned it instead of H200 SXM modules. They're trying to brag about cheap training. Maybe you wanted to use an opportunity to shill tranium. | | |
| ▲ | phs318u 14 minutes ago | parent [-] | | > Maybe you wanted to use an opportunity to shill tranium. This was unnecessary and against HN guidelines to engage in good faith. |
|
|
|