Daily Alpha · Substack
· Post-Market Alpha · by Buzzberg Research
SemiAnalysis said Anthropic runs production inference and RL rollouts on Trainium2 and Trainium3, and Fidenza Macro took profit on its oil long.
Anthropic runs production inference and RL rollouts on Trainium
SemiAnalysis says Trainium has proven itself in inference, with Anthropic running production inference and RL rollouts on Trainium2 and Trainium3, and argues it is now good for training too.
Historically Trainium was not the best fit for research training.
Source →Fidenza Macro took profit on its oil long
After writing about being short SPX and long oil, the author took profit on oil because the 30-minute, four-hour and daily charts are all overbought on RSI.
A macro trader is booking oil gains into the rally.
Source →Asymmetrical Bets called the AI selloff mostly sentiment, with CoreWeave the exception
The weekly recap recalls last week's argument that the macro outlook changed with the inflation prints but the stock-level damage was mostly sentiment driven, with CoreWeave the one exception where credit stress looked real.
It separates credit stress at CoreWeave from sentiment elsewhere in AI stocks.
Source →