Cui Tianyi: Claude Mods and DeepSeek Harness Reach the Same Goal via Different Paths

TRUMP co-founder Bill Zanker revealed at Token2049 that his team is developing a TRUMP-themed game and a TV show, and is also seeking to acquire a TRUMP-themed amusement park. Zanker noted that the team’s goal in advancing these projects is to boost TRUMP’s value and satisfy its token holders.

US President Donald Trump said in a local time statement on October 7 that he is "not very keen" on reaching a deal with Iran. Earlier, NBC News reported that Trump and his national security team are discussing the possibility of resuming large-scale military operations against Iran in the coming weeks; Axios added that potential actions could include "massive bombing" of Iran’s energy, infrastructure and nuclear facilities. Speaking at a campaign rally in San Antonio, Texas, Trump noted Iran is willing to offer "anything" to end the conflict, but he currently does not wish to strike a deal. He also said US Special Envoy for Middle East Steve Witkoff is pushing forward related negotiations and "doing a good job". Recent US-Iran talks remain marked by clear divisions. US Vice President JD Vance stated that Iran must drastically reduce its nuclear enrichment capacity to meet Washington’s demands for ending the war. For its part, Iran has criticized US officials for conflicting stances and insisted its right to uranium enrichment is non-negotiable. Meanwhile, Middle East energy markets continue to be impacted by the conflict. Kpler data shows that excluding Iran, crude oil exports via the Gulf, combined with shipments from Saudi Arabia and the UAE, have recovered to around 18.5 million barrels per day (bpd) – the pre-conflict level. However, oil prices remain elevated. Kpler analysts believe that even without a new diplomatic agreement, Middle East oil shipments could gradually recover through operational adjustments, though the process will be slower and uneven.

Analyst Murphy wrote that when BTC’s price falls, liquidity from market makers naturally tilts toward bids, while traders place orders at lower prices to wait for entry points. Data shows Binance’s spot order book’s 5% depth (total liquidity within a 5% range of the current price) hit $80 million at one point when BTC dropped to around $83,000. Relative to prior periods, this level is higher than during the September 10–16 correction and close to BTC’s sideways consolidation phase at its July bottom. Calculated at $83,000, buy orders within the 5% depth range extend roughly to $79,000, meaning substantial buy support exists between $83,000 and $79,000, which could lower the risk of the price being quickly breached. Meanwhile, the gap between taker buy and sell volumes on Binance rose steadily as the price fell, signaling more aggressive buying and fewer aggressive selling orders. Murphy noted this shows traders are not displaying significant bearish sentiment: not only is there ample passive buy support below, but aggressive buying interest has actually strengthened during the price decline.

Beating AI Flash News: According to market sources, OpenAI is investigating error issues with ChatGPT and GPTs services for users in the Asia-Pacific region.

Beating AI News reports that former ByteDance intern Tian Keyu has disclosed his startup project focused on world models. His team has raised nearly $30 million from investors including Source Code Capital and IDG Capital, with a post-money valuation of $200 million. The team currently has around 10 members and has not yet launched any public products. World models are designed to enable AI to learn how objects move and change in the real world, with applications in video generation, robotics, and other sectors. Tian, the first author of the best paper at NeurIPS 2024, previously researched how AI generates images gradually from coarse to fine. Building on this work, his new team has developed a "vocabulary" of 200,000 visual tokens for video—these operate similarly to the tokens used by large language models to process text, allowing AI to handle video in a more computationally efficient manner. The team plans to train its model on 100 million hours of video. Tian states this approach could cut the cost of generating one second of video to one-tenth or even lower than current levels. The team intends to hold a public demo first, with the full model scheduled for release in 2027. He also told Bloomberg that he once shut down another intern’s program, claiming it was to free up computing power for other researchers, and admitted the action was improper. ByteDance fired him in 2024 and subsequently sued for 8 million yuan. Court documents reviewed by Bloomberg show the court ordered Tian to pay 500,000 yuan for breach of contract.

Beating AI News – Third-party evaluation firm Artificial Analysis has released test results for Claude Haiku 5.5. The new model scored 43 points in the General Intelligence Index, a 26-point improvement over the previous generation’s 17 points. It outperformed GPT-6 Luna (38 points) and GLM-5.3 Flash (42 points), and is close to Kimi K3 (44 points). The index integrates 10 tests including programming, professional work, and scientific reasoning. However, the high score comes with substantial token consumption. At maximum inference intensity, Haiku 5.5 outputs an average of ~162,000 tokens per evaluation task, 3.2 times that of GPT-6 Luna. Both models have the same base API unit price, but Haiku 5.5’s estimated per-task cost is ~$0.21, compared to Luna’s $0.07. When adjusted from the second-highest tier to the maximum tier, Haiku 5.5’s overall score only rises by 2 points, while its output volume increases by ~80%. Reducing inference intensity narrows the gap significantly: Haiku 5.5 scores 38 points at the high tier, with an average of ~55,000 tokens per task, matching GPT-6 Luna’s 38 points and 50,000 tokens at the maximum tier. Developers can adjust inference intensity to cut unnecessary token consumption. Haiku 5.5 is not fully superior, though. In the AutomationBench-AA automation task evaluation, it scored only 35%, lagging behind Luna’s 53%. However, the model had an issue of over-refusing tasks during testing; Anthropic is addressing this, and Artificial Analysis plans to retest. Additionally, the currently disclosed task cost for Haiku 5.5 does not include long context surcharges, so actual fees may be higher.