Cointime

Download App
iOS & Android

Wall Street Rating Summary: UBS Raises AMD Target Price to $670, Seaport Maintains Sell Rating on Nvidia

On June 24, several Wall Street institutions released important rating adjustments involving companies such as Nvidia, Tesla, AMD, Reddit, and Arm. Seaport reiterated its 'Sell' rating on Nvidia, citing increasing burdens on Nvidia's balance sheet and the trend of Broadcom joining in on revolving financing, indicating that the AI industry is facing rising financing constraints. Stifel initiated coverage on BlackBerry with a 'Buy' rating, stating that the market continues to misdefine BlackBerry, which is no longer just an automotive supplier but a key software layer in the physical AI stack, collaborating with chip companies like Nvidia, Qualcomm, and AMD. UBS reaffirmed its 'Buy' rating on AMD and raised the target price from $470 to $670, expressing a more positive outlook as independent CPU racks begin to gain market recognition. TD Cowen reiterated its 'Buy' rating on Arm and raised the target price from $265 to $475, stating that Arm's $15 billion AGI CPU target for fiscal year 2031 is reasonable, although the GPU and CPU pairing rate and single-core pricing remain key variables. RBC reaffirmed its 'Outperform' rating on Tesla, expecting Tesla's second-quarter delivery volume to reach 405,000 vehicles, exceeding the Visible Alpha consensus estimate of 401,000 vehicles. Needham reiterated its 'Buy' rating on Reddit, noting that Reddit possesses a rare competitive position, largely unaffected by the proxy AI storm. Additionally, Citizens initiated coverage on Bitdeer, CleanSpark, Keel Infrastructure, and MARA Holdings, all giving them 'Outperform' ratings, stating that these Bitcoin mining companies are repurposing their existing power capacity to provide high-performance computing capabilities to hyperscale customers and large enterprises.

Comments

All Comments

Recommended for you

  • President Trump: Republicans Should Include Iran in Sanctions Bill Against Russia

    On July 19, President Trump stated that Republicans should include Iran in the sanctions bill aimed at Russia.
  • KPMG China Chairman: AI is Accelerating to Become the Core Engine of Economic Growth

    On July 19, KPMG China Chairman Zou Jun stated at the 2026 World Artificial Intelligence Conference that artificial intelligence has currently permeated various sectors of the real economy, rapidly transforming from a tool for enhancing efficiency to the core engine of economic growth. China's AI industry is actively positioned with diverse tracks and possesses supply chain advantages, leading to a rapid growth in the scale of related industries. A recent report released by KPMG during the conference pointed out that China has the most comprehensive industrial system in the world and a massive consumer market with over 1 billion internet users. The substantial industrial demand and consumption upgrades enable AI technology to achieve faster value transformation from laboratory to production line in the Chinese market. In terms of financing, KPMG data shows that embodied intelligence, as a major development direction for large model applications, has experienced rapid growth since 2025. The financing amount for embodied intelligence reached $11.17 billion in 2025, a year-on-year increase of 152.0%; the number of financing rounds was 670, up 81% year-on-year. In the first quarter of 2026, the financing amount reached $5.03 billion, a year-on-year increase of 182.9%; the number of financing rounds was 203, up 55% year-on-year. (The Paper)
  • Tonghuashun: Expected Net Profit Increase of 75%-95% Year-on-Year for First Half of 2026

    On July 19, Tonghuashun (300033.SZ) announced that it expects the net profit attributable to shareholders of the listed company for the first half of 2026 to be between 878 million and 979 million yuan, representing a year-on-year increase of 75%-95%. In the first half of 2026, the demand for financial information services rapidly grew due to factors such as active trading in the capital market. During the reporting period, the company continued to promote the deep integration of artificial intelligence technology with financial information service operations, accelerating the application of large models and AI technology in financial intelligence, data terminals, and user service scenarios. This has continuously improved the intelligence level of products and service capabilities, further optimizing user experience and enhancing the company's core competitiveness. Based on a net profit of 256 million yuan in Q1, the expected net profit for Q2 is estimated to be between 622 million and 723 million yuan, indicating a quarter-on-quarter growth of 143%-182%.
  • DeepSeek V4 'Full Version' Exposed, Expected Release Tomorrow

    According to market news, the official version of DeepSeek V4 is expected to be released as soon as tomorrow. Currently, some individuals have already gained early access to the gray testing of DeepSeek V4 (GA). There are two versions in total: DeepSeek V4 Flash and DeepSeek V4 Pro. Blogger AiBattle provided a 'verification mantra': look at the first-person perspective of the thought chain (CoT). If the model's thought process starts with 'I'm' or 'I'll', rather than the older version's 'Let me', it indicates a significant upgrade. Its performance is close to Opus 4.8, with encoding nearly matching GPT-5.6Sol, but at a lower price. The API introduces peak and off-peak billing, with peak prices doubling and cache hit rates being extremely low. While performance may not be the absolute best, the pricing remains aggressive, bringing Opus-level capabilities down to one-seventh the cost. (Source: New Intelligence)
  • Release of Alibaba Token Plan Personal Edition with Qwen3.8-Max Preview

    Just now, Alibaba officially launched the personal edition of the Token Plan, allowing users to get an early experience of Qwen3.8-Max-Preview. Qwen3.8-Max has a parameter count of 2.4T, demonstrating outstanding performance in scenarios such as code engineering and professional office work, making it potentially the most powerful model after Fable 5. The official version will be released and open-sourced soon. The Token Plan offers limited-time discounts: Personal Edition: Lite at 39 yuan/month (35% off), Standard at 139 yuan/month (23% off), Pro at 499 yuan/month (17% off). Team Edition: Standard seat at 150 yuan/seat/month (24% off), Premium seat at 550 yuan/seat/month (21% off), Exclusive seat at 1398 yuan/seat/month. Limited-time promotion for Qwen3.8-Max-Preview: Daytime credits consumption at 10% off, with an additional 20% off for personal edition Token Plan at night. In addition to the Token Plan, Qoder and QoderWork also support Qwen3.8-Max-Preview, and Qianwen PC users can experience it for free. (Wall Street Journal)
  • Musk Claims His 2T Model May Surpass Kimi, Initial Training to Complete Next Week

    On July 19, the new generation open-source large model Kimi K3 was released by Moon's Dark Side. According to reports, Tesla CEO Elon Musk commented in the evaluation report section, saying, 'Impressive.' Subsequently, on July 18, Musk revealed on the social media platform 'X' that his 2T parameter model will complete its initial training next week, and the new model will significantly outperform the 1.5T version. He also mentioned that this model may be able to surpass Kimi.
  • Trump Confirms Attendance at World Cup Final, U.S. Initiates Highest Level of Security

    On July 19, it was announced that the World Cup final between the U.S., Canada, and Mexico will take place at 3 AM on July 20 at the New Jersey Sports Complex in New York. President Trump has confirmed his attendance, and the U.S. has initiated a 'Level One' special security operation. F-16 fighter jets, military snipers, and thousands of FBI agents will be deployed on-site, with temporary flight restrictions and no-fly zone controls implemented in the New York and New Jersey areas. On the day of the event, traffic control will be enforced around the stadium, and fans are required to arrive four hours before the start of the match. Media must complete entry and security checks by 12:30 PM Eastern Time on July 19.
  • Chinese Academy of Engineering Academician Zheng Weimin: The Ability to Produce Stable and Low-Cost Tokens is Rare in the Era of Intelligent Agents

    On July 19, Zheng Weimin, an academician of the Chinese Academy of Engineering and a professor in the Department of Computer Science and Technology at Tsinghua University, stated at the AI Infrastructure Innovation Development Forum during the Agentic Era that the rapid expansion of computing power does not equate to the generation of efficient token production capacity. What is truly scarce in the era of intelligent agents is not just chips, but the system capability to produce tokens stably, at low cost, and at a high level. Tokens are transitioning from being a technical metric to becoming a core production factor in the industry, and reasoning systems are evolving from single-machine optimization to distributed, cached, heterogeneous, and service-oriented architectures. (Securities Times)
  • Cleveland Fed President Hints at Rate Hike to Combat Inflation, Market Bets on 65% Chance of September Increase

    On July 19, Cleveland Fed President Loretta Mester suggested that the Federal Reserve may need to raise interest rates to curb persistently high inflation. Currently, futures traders estimate the probability of a rate hike in July at about 15%, while the probability for September stands at approximately 65%.
  • Alibaba Qwen: Qwen3.8 Set to Release with Open Weights

    Alibaba Qwen announced that Qwen3.8 is set to be released with open weights. This model boasts a parameter scale of up to 2.4 trillion and is continuously evolving. We believe it is one of the most powerful models currently available, performing comparably to leading cutting-edge AI models, second only to Fable 5. Just recently, Qwen3.8-Max-Preview made its debut and has been launched on Alibaba's Token Plan, Qoder, and QoderWork platforms. (Qwen Large Model WeChat Official Account)