▲ 8 ▼ DeepSeek V4 Pro at 5% the cost of Claude — what it takes to close the gap (howardchen.substack.com) submitted 4 months ago by cm0002@mander.xyz to c/Aii@programming.dev 8 comments fedilink hide all child comments
[–] DacoTaco@lemmy.world 2 points 4 months ago (6 children) Tbf, qwen is a distilled version of claude :') permalink fedilink source parent hideshow 6 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (5 children) Really? What do you mean? permalink fedilink source parent hideshow 5 child comments replies: [–] jacksilver@lemmy.world 1 point 4 months ago (4 children) Its believed (probably more just fact) that a lot of the open source models leverage larger models. Either directly through distiallation (meaning training a model directly off of another) or by using the larger models for knowledge distiallation (basically using the larger models as "ground truth" for training workflows). Either way, this means the open source models would always be "a few months behind" as they rely on the progress of the closed source models to improve themselves. permalink fedilink source parent hideshow 4 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (3 children) I see. Is there some way to protect models from being trained on them aside from just removing it from the market? permalink fedilink source parent hideshow 3 child comments replies: [–] jacksilver@lemmy.world 1 point 4 months ago (2 children) From what I've read, it seems like a lot of the companies monitor API calls and will posion responses for suspicious traffic. This basically means they'll modify the output to be harmful to the scrapper. Outside of that they can block accounts/IPs, limit what is returned (just the final answer vs all reasoning steps, etc.). However it's a bit like all cyber security and an evolving game of cat and mouse. permalink fedilink source parent hideshow 2 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (1 child) How was Qwen previously able to train on Claude? Prompt injection? permalink fedilink source parent hideshow 1 child comment replies: [–] jacksilver@lemmy.world 1 point 4 months ago I'm not caught up on all the details, but as I said this has been an ongoing game of cat and mouse with people looking to leverage other models to improve their own (this isn't just a qwen vs Claude thing and everyone if pointing fingers at everyone). If I had to guess it's only been more recently that AI companies are actively fighting these issues, but even then I imagine they're not 100% effective. permalink fedilink source parent
[–] venusaur@lemmy.world 1 point 4 months ago (5 children) Really? What do you mean? permalink fedilink source parent hideshow 5 child comments replies: [–] jacksilver@lemmy.world 1 point 4 months ago (4 children) Its believed (probably more just fact) that a lot of the open source models leverage larger models. Either directly through distiallation (meaning training a model directly off of another) or by using the larger models for knowledge distiallation (basically using the larger models as "ground truth" for training workflows). Either way, this means the open source models would always be "a few months behind" as they rely on the progress of the closed source models to improve themselves. permalink fedilink source parent hideshow 4 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (3 children) I see. Is there some way to protect models from being trained on them aside from just removing it from the market? permalink fedilink source parent hideshow 3 child comments replies: [–] jacksilver@lemmy.world 1 point 4 months ago (2 children) From what I've read, it seems like a lot of the companies monitor API calls and will posion responses for suspicious traffic. This basically means they'll modify the output to be harmful to the scrapper. Outside of that they can block accounts/IPs, limit what is returned (just the final answer vs all reasoning steps, etc.). However it's a bit like all cyber security and an evolving game of cat and mouse. permalink fedilink source parent hideshow 2 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (1 child) How was Qwen previously able to train on Claude? Prompt injection? permalink fedilink source parent hideshow 1 child comment replies: [–] jacksilver@lemmy.world 1 point 4 months ago I'm not caught up on all the details, but as I said this has been an ongoing game of cat and mouse with people looking to leverage other models to improve their own (this isn't just a qwen vs Claude thing and everyone if pointing fingers at everyone). If I had to guess it's only been more recently that AI companies are actively fighting these issues, but even then I imagine they're not 100% effective. permalink fedilink source parent
[–] jacksilver@lemmy.world 1 point 4 months ago (4 children) Its believed (probably more just fact) that a lot of the open source models leverage larger models. Either directly through distiallation (meaning training a model directly off of another) or by using the larger models for knowledge distiallation (basically using the larger models as "ground truth" for training workflows). Either way, this means the open source models would always be "a few months behind" as they rely on the progress of the closed source models to improve themselves. permalink fedilink source parent hideshow 4 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (3 children) I see. Is there some way to protect models from being trained on them aside from just removing it from the market? permalink fedilink source parent hideshow 3 child comments replies: [–] jacksilver@lemmy.world 1 point 4 months ago (2 children) From what I've read, it seems like a lot of the companies monitor API calls and will posion responses for suspicious traffic. This basically means they'll modify the output to be harmful to the scrapper. Outside of that they can block accounts/IPs, limit what is returned (just the final answer vs all reasoning steps, etc.). However it's a bit like all cyber security and an evolving game of cat and mouse. permalink fedilink source parent hideshow 2 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (1 child) How was Qwen previously able to train on Claude? Prompt injection? permalink fedilink source parent hideshow 1 child comment replies: [–] jacksilver@lemmy.world 1 point 4 months ago I'm not caught up on all the details, but as I said this has been an ongoing game of cat and mouse with people looking to leverage other models to improve their own (this isn't just a qwen vs Claude thing and everyone if pointing fingers at everyone). If I had to guess it's only been more recently that AI companies are actively fighting these issues, but even then I imagine they're not 100% effective. permalink fedilink source parent
[–] venusaur@lemmy.world 1 point 4 months ago (3 children) I see. Is there some way to protect models from being trained on them aside from just removing it from the market? permalink fedilink source parent hideshow 3 child comments replies: [–] jacksilver@lemmy.world 1 point 4 months ago (2 children) From what I've read, it seems like a lot of the companies monitor API calls and will posion responses for suspicious traffic. This basically means they'll modify the output to be harmful to the scrapper. Outside of that they can block accounts/IPs, limit what is returned (just the final answer vs all reasoning steps, etc.). However it's a bit like all cyber security and an evolving game of cat and mouse. permalink fedilink source parent hideshow 2 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (1 child) How was Qwen previously able to train on Claude? Prompt injection? permalink fedilink source parent hideshow 1 child comment replies: [–] jacksilver@lemmy.world 1 point 4 months ago I'm not caught up on all the details, but as I said this has been an ongoing game of cat and mouse with people looking to leverage other models to improve their own (this isn't just a qwen vs Claude thing and everyone if pointing fingers at everyone). If I had to guess it's only been more recently that AI companies are actively fighting these issues, but even then I imagine they're not 100% effective. permalink fedilink source parent
[–] jacksilver@lemmy.world 1 point 4 months ago (2 children) From what I've read, it seems like a lot of the companies monitor API calls and will posion responses for suspicious traffic. This basically means they'll modify the output to be harmful to the scrapper. Outside of that they can block accounts/IPs, limit what is returned (just the final answer vs all reasoning steps, etc.). However it's a bit like all cyber security and an evolving game of cat and mouse. permalink fedilink source parent hideshow 2 child comments replies: [–] venusaur@lemmy.world 1 point 4 months ago (1 child) How was Qwen previously able to train on Claude? Prompt injection? permalink fedilink source parent hideshow 1 child comment replies: [–] jacksilver@lemmy.world 1 point 4 months ago I'm not caught up on all the details, but as I said this has been an ongoing game of cat and mouse with people looking to leverage other models to improve their own (this isn't just a qwen vs Claude thing and everyone if pointing fingers at everyone). If I had to guess it's only been more recently that AI companies are actively fighting these issues, but even then I imagine they're not 100% effective. permalink fedilink source parent
[–] venusaur@lemmy.world 1 point 4 months ago (1 child) How was Qwen previously able to train on Claude? Prompt injection? permalink fedilink source parent hideshow 1 child comment replies: [–] jacksilver@lemmy.world 1 point 4 months ago I'm not caught up on all the details, but as I said this has been an ongoing game of cat and mouse with people looking to leverage other models to improve their own (this isn't just a qwen vs Claude thing and everyone if pointing fingers at everyone). If I had to guess it's only been more recently that AI companies are actively fighting these issues, but even then I imagine they're not 100% effective. permalink fedilink source parent
[–] jacksilver@lemmy.world 1 point 4 months ago I'm not caught up on all the details, but as I said this has been an ongoing game of cat and mouse with people looking to leverage other models to improve their own (this isn't just a qwen vs Claude thing and everyone if pointing fingers at everyone). If I had to guess it's only been more recently that AI companies are actively fighting these issues, but even then I imagine they're not 100% effective. permalink fedilink source parent