Comment by s1artibartfast 10 days ago Users always have the choice to ignore benchmarks or run their own 4 comments s1artibartfast Reply danuker 10 days ago Exactly. Even a brief one (a single task done on two models) showed me GLM 5.3 Flash was superior to DeepSeek 4.1 Flash.DS used nearly twice the budget to accomplish the same thing.Ideally, these tasks should be something you actually need day to day. lisplist 9 days ago I'd rank it DeepSeek 4.1 Flash 1017 > GLM 5.3 Flash > DeepSeek 4.1 Flash (Original) - 1017 is a huge improvement. danuker 9 days ago Oh! I thought I had tested 1017, but I forgot about it. Thank you! adastra22 10 days ago That’s missing the point. Frontier models get shittier in certain ways with each release, due to benchmaxxing, and there is nothing we can do about that.
danuker 10 days ago Exactly. Even a brief one (a single task done on two models) showed me GLM 5.3 Flash was superior to DeepSeek 4.1 Flash.DS used nearly twice the budget to accomplish the same thing.Ideally, these tasks should be something you actually need day to day. lisplist 9 days ago I'd rank it DeepSeek 4.1 Flash 1017 > GLM 5.3 Flash > DeepSeek 4.1 Flash (Original) - 1017 is a huge improvement. danuker 9 days ago Oh! I thought I had tested 1017, but I forgot about it. Thank you!
lisplist 9 days ago I'd rank it DeepSeek 4.1 Flash 1017 > GLM 5.3 Flash > DeepSeek 4.1 Flash (Original) - 1017 is a huge improvement. danuker 9 days ago Oh! I thought I had tested 1017, but I forgot about it. Thank you!
adastra22 10 days ago That’s missing the point. Frontier models get shittier in certain ways with each release, due to benchmaxxing, and there is nothing we can do about that.
Exactly. Even a brief one (a single task done on two models) showed me GLM 5.3 Flash was superior to DeepSeek 4.1 Flash.
DS used nearly twice the budget to accomplish the same thing.
Ideally, these tasks should be something you actually need day to day.
I'd rank it DeepSeek 4.1 Flash 1017 > GLM 5.3 Flash > DeepSeek 4.1 Flash (Original) - 1017 is a huge improvement.
Oh! I thought I had tested 1017, but I forgot about it. Thank you!
That’s missing the point. Frontier models get shittier in certain ways with each release, due to benchmaxxing, and there is nothing we can do about that.