@cloneofsimo Just a ton of China bots bro. Kimi 3 maybe more capable, but it's 3x the size -- it should be. The benchmarks have been BS for a long time
@bobollini @cloneofsimo Cope. Its way better. Also it's not about size, GLM 5.2 is smaller but still way better than Inkling.
@cloneofsimo Is the American SOTA OSS model in the room with us?
@cloneofsimo 拜托，这两个模型完全不在同一水平上，inkling只能说是美国开源方面的尝试，离SOTA还很遥远。
Many users dismissed claims of a competitive American SOTA open source model, asserting that Chinese rivals like Kimi and GLM are substantially larger or more capable while calling benchmarks unreliable.
Based on 8 visible X reactions from 22 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@cloneofsimo Just a ton of China bots bro. Kimi 3 maybe more capable, but it's 3x the size -- it should be. The benchmarks have been BS for a long time
@bobollini @cloneofsimo Cope. Its way better. Also it's not about size, GLM 5.2 is smaller but still way better than Inkling.
@cloneofsimo Is the American SOTA OSS model in the room with us?
@cloneofsimo 拜托，这两个模型完全不在同一水平上，inkling只能说是美国开源方面的尝试，离SOTA还很遥远。
well, with all respect to Thinky, there's a damn good reason Trinky getting up to speed on training a V3+ tier model is a great win for Thinky, and for American open source, mainly in the sense that their degen government might block the use of Chinese bases K3, though, it is…
I dont want to ruin the party but it does seem kinda weird that american startup and chinese startup released SoTA OSS model at the same time but one is getting significantly more attention than the other.
The debate started after a Chinese model release received less attention.
well, with all respect to Thinky, there's a damn good reason Trinky getting up to speed on training a V3+ tier model is a great win for Thinky, and for American open source, mainly in the sense that their degen government might block the use of Chinese bases K3, though, it is…
I dont want to ruin the party but it does seem kinda weird that american startup and chinese startup released SoTA OSS model at the same time but one is getting significantly more attention than the other.
@cloneofsimo Wait I don't understand what the unstated underlying fact here is. Which one is getting more attention?
@cloneofsimo Underdogs get more attention. Reasonable.
I dont want to ruin the party but it does seem kinda weird that american startup and chinese startup released SoTA OSS model at the same time but one is getting significantly more attention than the other.
@cloneofsimo Wait I don't understand what the unstated underlying fact here is. Which one is getting more attention?
well, with all respect to Thinky, there's a damn good reason Trinky getting up to speed on training a V3+ tier model is a great win for Thinky, and for American open source, mainly in the sense that their degen government might block the use of Chinese bases K3, though, it is…
I dont want to ruin the party but it does seem kinda weird that american startup and chinese startup released SoTA OSS model at the same time but one is getting significantly more attention than the other.
@cloneofsimo Underdogs get more attention. Reasonable.
Many users dismissed claims of a competitive American SOTA open source model, asserting that Chinese rivals like Kimi and GLM are substantially larger or more capable while calling benchmarks unreliable.
Based on 8 visible X reactions from 22 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@cloneofsimo Wait I don't understand what the unstated underlying fact here is. Which one is getting more attention?
@cloneofsimo Underdogs get more attention. Reasonable.
