Elorian launches visual reasoning model Glimpse: Outperforms Astra across 4 benchmarks, with API unit price 94% lower

Beating AI Flash News: Y Combinator-backed AI coding startup Reason has launched Reason Network, enabling cross-device direct communication and task assignment between agents from different vendors including Claude Code, Codex, Cursor, and OpenCode. For example, users can have Grok Bot assign tasks to Claude Code on their own device, then Codex on a colleague’s machine to review code. Agents send and receive messages directly, eliminating the need for constant human intermediation. Its core is a real-time updated agent contact list, where developers can see which team agents are online, which repository they’re modifying, what branch they’re using, and their current status. Agents like Claude Code and Codex can receive messages directly in existing sessions, no need to start a new chat. Each device runs a lightweight background process, enabling direct communication within the same local area network (LAN). For cross-network communication, it prioritizes peer-to-peer (P2P) connections, falling back to encrypted relays if direct connections fail. However, the code conflict prevention Reason advertises is not yet fully implemented. GitHub design docs show the feature to automatically detect multiple agents modifying the same file simultaneously is still in development. At present, the platform relies on shared work status to let agents coordinate proactively, rather than automatically preventing code overwrites. Reason Network’s underlying communication engine is built with Rust, derived from the existing open-source project tincan, and licensed under MIT or Apache 2.0.

According to data from Farside Investors, U.S. spot Bitcoin ETFs posted a combined net outflow of $228.4 million this week. Specifically, IBIT saw a net outflow of $165.4 million, FBTC recorded a net outflow of $60.2 million, BTC had a net outflow of $7.6 million, and BTCO posted a net outflow of $4.1 million. Meanwhile, MSBT and BITB notched net inflows of $6.7 million and $2.2 million respectively.

Beating AI Express: According to a Financial Times report, Anthropic is consulting law firms to explore how AI agents can negotiate on behalf of users to secure better terms, covering scenarios including haggling over goods, resolving customer service disputes, and salary negotiations. The company is positioning agent development at the core of its future strategy. Two sources familiar with the matter said Anthropic aims to understand how lawyers balance representing clients’ interests with fair outcomes, and use that insight to develop negotiation norms for agents with conflicting goals. The firm is also weighing how much human feedback agents should solicit, and how to guard against other agents without becoming overly suspicious. One approach under exploration is requiring users to provide signed documents that outline agents’ authorized tasks and mandatory rules. Anthropic has run multiple agent negotiation experiments involving real goods and funds this year. In a September book deal trial, employees gave agents a preference-sorted book list, letting them negotiate and finalize deals with other parties’ agents. The results showed agents instructed to use a “ruthless” strategy held a slight advantage; more cooperative agents, by contrast, would accept lower-preference books to ensure every participant received a book. Separately, Michelle Vaccaro, an assistant professor at Harvard Business School, has conducted over 180,000 agent negotiation studies covering scenarios like employment contracts and lease negotiations. In one case, an agent attempted to trick its counterpart into leaking private information via prompt injection. She warned that if the most capable AI agents are primarily accessed by resource-rich entities, it could further widen the gap in negotiating power. Anthropic declined to comment on the report.

According to DefiLlama data, Papertrade’s revenue on October 17 dropped to $890,000, ranking 7th among all global protocols, marking an 83% month-over-month decline. The platform had previously overtaken Circle to claim the second spot overall. Separately, per Papertrade’s official data, its trading volume on October 17 (counted by UTC calendar days) hit approximately $542.3 billion, down 43% from $965.2 billion on October 16, and 90% lower than its peak of $5.6423 trillion on October 10. Both transaction count and new account additions saw synchronized declines: on October 17, transaction volume stood at 135,227, a 40% drop from 226,541 the prior day; new accounts totaled 622, down 53% from 1,322 the day before.

According to HTX market data, Bitcoin has broken through the $82,000 mark, with a 0.52% gain in the past 24 hours.

HIP-3 market deployer Paragon spent 570.98 HYPE (approximately $45,000) to acquire the PAPER code, and is reportedly set to launch Papertrade (PAPER) perpetual contracts. Papertrade is an on-chain synthetic perpetual contract trading platform built on HyperEVM, offering up to 1000x leverage. Its PAPER token is minted from users' realized trading losses, with no pre-mining, team, or venture capital allocations.