Daily Alpha · Substack
· Post-Market Alpha · by Buzzberg Research
The strongest public items highlighted improving open-model utility, a Marvell earnings catalyst and an explicit rotation from leveraged energy into lithium and batteries.
Themes on this desk
Open-model utility
SemiAnalysis says GLM 5.3 and Kimi K3 now perform economically useful coding and agentic work, narrowing the practical gap with closed models.
Marvell catalyst
Damnang remains constructive on FY28 upside ahead of Marvell's August 27 earnings report.
Portfolio rotation
Rooster Global disclosed selling leveraged energy exposure and buying a lithium and battery fund.
Shift in open-source AI model utility
The author argues that recent open-source models like GLM 5.3 and Kimi K3 are achieving genuine utility in coding and agentic tasks, contrasting this with earlier models like DeepSeek R1 which they claim lacked economically valuable application.
Suggests a potential narrowing of the competitive gap between open-source and closed-source frontier AI models.
Watch Adoption rates of GLM 5.3 and Kimi K3 in enterprise or developer workflows.
Source →Marvell Technology FY28 upside thesis
The author maintains a positive view on Marvell Technology and anticipates upside for FY28 ahead of the August 27 earnings print.
Signals potential market anticipation of positive guidance or performance metrics for Marvell.
Watch Market reaction to the August 27 earnings report.
Source →Portfolio rotation: Selling NRGU and buying LITP
The author of the Rooster Global Portfolio Mastery Club newsletter disclosed a portfolio adjustment involving the sale of the Direxion Daily Energy Bull 3X Shares (NRGU) and the purchase of the Global X Lithium & Battery Tech ETF (LITP).
Indicates a tactical shift from leveraged energy exposure toward the battery and lithium supply chain sector.
Watch Further commentary on energy vs. battery tech sector outlooks from the author.
Source →