TFW you spend a few hours struggling to get American open weight models working on various clouds while Kimi and DeepSeek are plug and play
3:21 AM · May 30, 2026 · 16K Views
AVERI's Miles Brundage says deploying US open-weight models took hours compared to seamless Chinese alternatives
Florian Brand suggested OpenRouter to simplify the deployment process.
1519741426.9K
Sentiment
Sentiment building, check back later.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Related links
Posts from X
Most Activity
Most Activity
VIEWS2.8KBOOKMARKS3LIKES38REPLIES2
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex
Btw I think this era will last only one more year. Western open models will standardize around Nemotrons (and similar designs) and Nvidia will make sure they run on neoclouds. And Chinese ones… will start to diverge. Ascend-shaped models probably won't be very Nvidia-friendly.
Miles Brundage@Miles_Brundage
TFW you spend a few hours struggling to get American open weight models working on various clouds while Kimi and DeepSeek are plug and play
30dViews 2.8KLikes 38Bookmarks 3
RETWEETS1
Miles Brundage@Miles_Brundage
(earlier version of this tweet specified open source which really has been the main pain point, e.g. getting OLMo and Marin variants going, + should have kept that in in retrospect - there are American options but there tends to be a performance penalty and/or they're opaque)
Miles Brundage@Miles_Brundage
TFW you spend a few hours struggling to get American open weight models working on various clouds while Kimi and DeepSeek are plug and play
30dViews 2.7KLikes 10Bookmarks 0
Florian Brand@xeophon
@Miles_Brundage which one, openrouter has a ton of them?
Miles Brundage@Miles_Brundage
TFW you spend a few hours struggling to get American open weight models working on various clouds while Kimi and DeepSeek are plug and play
30dViews 1.7KLikes 17Bookmarks 0
Miles Brundage@Miles_Brundage
@xeophon
Miles Brundage@Miles_Brundage
(earlier version of this tweet specified open source which really has been the main pain point, e.g. getting OLMo and Marin variants going, + should have kept that in in retrospect - there are American options but there tends to be a performance penalty and/or they're opaque)
30dViews 825Likes 5Bookmarks 1
Miles Brundage@Miles_Brundage
(cc @natolambert )
Miles Brundage@Miles_Brundage
TFW you spend a few hours struggling to get American open weight models working on various clouds while Kimi and DeepSeek are plug and play
30dViews 2.6KLikes 9Bookmarks 0
Kyle Wiggers@Kyle_L_Wiggers
@Miles_Brundage check out our partner cirrascale's apis https://www.cirrascale.com/ai2endpoints
Miles Brundage@Miles_Brundage
(earlier version of this tweet specified open source which really has been the main pain point, e.g. getting OLMo and Marin variants going, + should have kept that in in retrospect - there are American options but there tends to be a performance penalty and/or they're opaque)
30dViews 137Likes 1Bookmarks 0
Florian Brand@xeophon
@Miles_Brundage is it as simple as kimi? no. but also the more apt. comparison would be something like openbmb or internlm
Florian Brand@xeophon
@Miles_Brundage hm, at least if you don't need max performance, renting a gpu and asking claude/codex to set up vllm and the correct parser etc is what i do and will give you an endpoint in like 10-20 minutes
30dViews 147Likes 3Bookmarks 0
Florian Brand@xeophon
@Miles_Brundage hm, at least if you don't need max performance, renting a gpu and asking claude/codex to set up vllm and the correct parser etc is what i do and will give you an endpoint in like 10-20 minutes
30dViews 9
Kyle Wiggers@Kyle_L_Wiggers
@Miles_Brundage we're working on it!!
Miles Brundage@Miles_Brundage
@Kyle_L_Wiggers (going w/ self hosting on Parasail though wish there were more versions available there)
30dViews 37Likes 1Bookmarks 0