Search “Agents”

Nathan Lambert says he doesn't know how to try out Gemini in the agent era

On October 29, 2026, Nathan Lambert said he didn't know how to try out Gemini in the agent era and believed he needed to make up for this gap. He mentioned that Gemini seemed to have been absent since December 2025, but might be making a comeback. The tweet quoted Sundar Pichai's post announcing Gemini 4 Carbon, which stated that the model ranked first with 59 points on the Artificial Analysis Intelligence Index v5.0, and scored 88.2% on SEC-bench Pro, 54.8% on ExploitGym, and 71.0% on CWE-bench. There have been no further developments so far.

Nathan Lambert says he doesn't know how to try Gemini in the Agent era

On October 29, 2026, Nathan Lambert said he didn't know how to try out Gemini in the agent era and believed he needed to make up for this gap. He mentioned that Gemini seemed to have been absent since December 2025, but might be making a comeback. The tweet quoted Sundar Pichai's post announcing Gemini 4 Carbon, which stated that the model ranked first with 59 points on the Artificial Analysis Intelligence Index v5.0, and scored 88.2% on SEC-bench Pro, 54.8% on ExploitGym, and 71.0% on CWE-bench. There have been no further developments so far.

View story:Nathan Lambert says he doesn't know how to try out Gemini in the agent era

Alibaba Cloud open-sources the "AI Agent Handbook"

On October 30, 2026, Alibaba Cloud open-sourced the "AI Agent Handbook," an enterprise-grade Agent full-lifecycle handbook covering five parts—architecture, building, operation, governance, and tuning—with Harness as its core, aiming to move enterprise Agents from pilot to governable production. A 2026 survey cited in the handbook shows that enterprises that have completed or are in the process of development account for about 46%, while those actually deployed to production account for about 18%; context and memory needs account for about 90%, multi-model routing and cost governance about 63%, and trajectory-based automated evaluation less than about 8%. On October 31, the "2026 Agent Developer Survey Report" included in the handbook further shows that teams engaged in continuous iteration account for less than 10%; for evaluation, manual spot checks account for about 55% and automated evaluation based on runtime trajectories about 6%–8%; architecturally, single-Agent setups account for about 40% and multi-Agent about 42%.

Alibaba Cloud open-sources "AI Agent Handbook"

On October 30, 2026, Alibaba Cloud open-sourced the "AI Agent Handbook," an enterprise-grade Agent full-lifecycle handbook covering five parts—architecture, building, operation, governance, and tuning—with Harness as its core, aiming to move enterprise Agents from pilot to governable production. A 2026 survey cited in the handbook shows that enterprises that have completed or are in the process of development account for about 46%, while those actually deployed to production account for about 18%; context and memory needs account for about 90%, multi-model routing and cost governance about 63%, and trajectory-based automated evaluation less than about 8%. On October 31, the "2026 Agent Developer Survey Report" included in the handbook further shows that teams engaged in continuous iteration account for less than 10%; for evaluation, manual spot checks account for about 55% and automated evaluation based on runtime trajectories about 6%–8%; architecturally, single-Agent setups account for about 40% and multi-Agent about 42%.

View story:Alibaba Cloud open-sources the "AI Agent Handbook"

Matt Palmer demonstrated using the Cursor cloud Agent to generate an interface motion animation video against real Slack

On October 30, 2026, Matt Palmer demonstrated opening Chrome and real Slack on the Linux remote desktop of the Cursor cloud Agent, having the Agent use code to generate a Slack interface motion animation video, and then iterating and revising after checking the easing and motion patterns against the real interface. He broke the process down into an inner loop (the Agent does the work, checks against the real thing, and fixes it) and an outer loop (a human provides references, preferences, and permissions and accepts the output), and proposed the ladder of Do→Codify→Generalize, turning one success into a Skill and then generalizing it. He emphasized that Slack was not decompiled; he merely gave the Agent a reference and asked it to extract visual patterns by comparison.

Matt demonstrates using Cursor Cloud Agent to generate UI motion videos by referencing real Slack

On October 30, 2026, Matt Palmer demonstrated opening Chrome and real Slack on the Linux remote desktop of the Cursor cloud Agent, having the Agent use code to generate a Slack interface motion animation video, and then iterating and revising after checking the easing and motion patterns against the real interface. He broke the process down into an inner loop (the Agent does the work, checks against the real thing, and fixes it) and an outer loop (a human provides references, preferences, and permissions and accepts the output), and proposed the ladder of Do→Codify→Generalize, turning one success into a Skill and then generalizing it. He emphasized that Slack was not decompiled; he merely gave the Agent a reference and asked it to extract visual patterns by comparison.

View story:Matt Palmer demonstrated using the Cursor cloud Agent to generate an interface motion animation video against real Slack

TestingCatalog says Google has terminated the Jules AI agent

On October 30, 2026, TestingCatalog reported that Google will terminate its AI coding agent Jules. Phoenix Technology reported the same day that Google will gradually shut down Jules, which was launched in December 2024, and will integrate the experience accumulated from the tool into its AI coding platform Antigravity; Jules' official X account stated that Jules was originally an experiment exploring how asynchronous coding agents could help developers ship faster, and that its experience has already influenced multiple Google products. cnBeta reported the same day that Jules will gradually cease operations, less than two years after its launch, and that the team is integrating the experience into Antigravity so that users can both autonomously execute background tasks in remote environments and use an agent-centric development environment. The current status is gradual shutdown and experience integration.

TestingCatalog says Google terminates Jules AI agent

On October 30, 2026, TestingCatalog reported that Google will terminate its AI coding agent Jules. Phoenix Technology reported the same day that Google will gradually shut down Jules, which was launched in December 2024, and will integrate the experience accumulated from the tool into its AI coding platform Antigravity; Jules' official X account stated that Jules was originally an experiment exploring how asynchronous coding agents could help developers ship faster, and that its experience has already influenced multiple Google products. cnBeta reported the same day that Jules will gradually cease operations, less than two years after its launch, and that the team is integrating the experience into Antigravity so that users can both autonomously execute background tasks in remote environments and use an agent-centric development environment. The current status is gradual shutdown and experience integration.

View story:TestingCatalog says Google has terminated the Jules AI agent

Shopify CEO Tobi discusses how personal AI Agents can make money

Shopify CEO Tobi Lütke offered his judgment on how personal AI agents will monetize. He said personal AI agents will not monetize through a 2% transaction cut, but will enter an era of "basis points." He also proposed a "reverse money test": users would be willing to pay an increasingly high price to stop using their own Muse. The monetization method could be metered billing in the background, or a cut of transaction volume, or both; he is not worried, because there is a lot of room for agents to help people save money.

Shopify CEO Tobi discusses how personal AI agents can make money

Shopify CEO Tobi Lütke offered his judgment on how personal AI agents will monetize. He said personal AI agents will not monetize through a 2% transaction cut, but will enter an era of "basis points." He also proposed a "reverse money test": users would be willing to pay an increasingly high price to stop using their own Muse. The monetization method could be metered billing in the background, or a cut of transaction volume, or both; he is not worried, because there is a lot of room for agents to help people save money.

View story:Shopify CEO Tobi discusses how personal AI Agents can make money

Tripo launches beta version of Agentic Canvas in Tripo Studio

On October 29, 2026, Tripo announced that Agentic Canvas in Tripo Studio has entered the beta phase, opening it to all MAX and Team users, with a gradual rollout to other users. This feature allows users to describe their needs in natural language, with an agent planning and executing the workflow to directly produce 3D assets, supporting batch processing and reusable pipelines, and enabling asset inspection, refinement, rigging, and editing through AI 3D Lab. There is currently no information on further developments.

Meshy demonstrates Hermes Agent generating and reskinning a claw machine game in a single conversation

On October 29, 2026, Meshy demonstrated an achievement: building and reskinning a claw machine game within a single Hermes Agent conversation. According to information released by Meshy, Meshy generated all 12 game assets, and Claude Code built the game around these assets; each replay round completely changes the art style, in order: plush, gummy, and pixel art. The same game presents three art styles without manual modeling. Currently, this event only has release content from Meshy as a single source, and there are no subsequent developments.

Arena.ai compares coding agent user criticism across 20,869 traces

Based on 20,869 traces from Agent Arena: Code, Arena.ai compared user criticism signals of five top models from Anthropic and OpenAI against other agents. The analysis found that frontier models differ most noticeably in three dimensions: 6.9 percentage points fewer complaints about failure-fix loops and 1.9 percentage points fewer about ignoring project context; 6.4 percentage points fewer design issues and 5.8 percentage points fewer issues with producing usable results; and 5.6, 4.9, and 4.5 percentage points fewer complaints about instruction following, task completion, and breaking existing code, respectively. There are currently no follow-up developments.

OpenBMB says its Agent approached frontier level within four hours

On October 29, 2026, ModelBest OpenBMB announced that its agent, starting from scratch in just four hours, has reached performance very close to the frontier level. The post also cited an earlier post about ForgeKernel Live optimizing the DeepSeek kernel. At present, this claim appears only in content published by OpenBMB itself, and there are no independent reports or follow-up developments from other sources.

Arena says GPT-6.1 Terra (Max) ranks 8th in Agent Arena

Arena released measured results based on nearly 9,000 agent sessions: OpenAI's GPT-6.1 Terra (Max) ranks 8th in Agent Arena, with a net improvement of +10.6% and a median cost per task of $0.62. This cost is lower than the $1.09 of 7th-place Gemini 4 Argon (High), the $0.89 of 6th-place GPT-6.1 Sol (Max), and the $5.16 of 5th-place GPT-6 Astra (Max). In the subcategory rankings, it places 3rd in Bash Reliability (+13.2%), 7th in Steerability (+13.3%), and 9th in Praise vs Complaint (+16.8%). There are currently no further developments.

Tripo launches beta version of Agentic Canvas in Tripo Studio

On October 29, 2026, Tripo announced that Agentic Canvas in Tripo Studio has entered the beta phase, opening it to all MAX and Team users, with a gradual rollout to other users. This feature allows users to describe their needs in natural language, with an agent planning and executing the workflow to directly produce 3D assets, supporting batch processing and reusable pipelines, and enabling asset inspection, refinement, rigging, and editing through AI 3D Lab. There is currently no information on further developments.

View story:Tripo launches beta version of Agentic Canvas in Tripo Studio

Meshy showcases Hermes Agent generating and reskinning a claw machine game in a single conversation

On October 29, 2026, Meshy demonstrated an achievement: building and reskinning a claw machine game within a single Hermes Agent conversation. According to information released by Meshy, Meshy generated all 12 game assets, and Claude Code built the game around these assets; each replay round completely changes the art style, in order: plush, gummy, and pixel art. The same game presents three art styles without manual modeling. Currently, this event only has release content from Meshy as a single source, and there are no subsequent developments.

View story:Meshy demonstrates Hermes Agent generating and reskinning a claw machine game in a single conversation

OpenBMB says its Agent approaches frontier level within four hours

On October 29, 2026, ModelBest OpenBMB announced that its agent, starting from scratch in just four hours, has reached performance very close to the frontier level. The post also cited an earlier post about ForgeKernel Live optimizing the DeepSeek kernel. At present, this claim appears only in content published by OpenBMB itself, and there are no independent reports or follow-up developments from other sources.

View story:OpenBMB says its Agent approached frontier level within four hours

Arena says GPT-6.1 Terra (Max) ranks 8th on Agent Arena

Arena released measured results based on nearly 9,000 agent sessions: OpenAI's GPT-6.1 Terra (Max) ranks 8th in Agent Arena, with a net improvement of +10.6% and a median cost per task of $0.62. This cost is lower than the $1.09 of 7th-place Gemini 4 Argon (High), the $0.89 of 6th-place GPT-6.1 Sol (Max), and the $5.16 of 5th-place GPT-6 Astra (Max). In the subcategory rankings, it places 3rd in Bash Reliability (+13.2%), 7th in Steerability (+13.3%), and 9th in Praise vs Complaint (+16.8%). There are currently no further developments.

View story:Arena says GPT-6.1 Terra (Max) ranks 8th in Agent Arena

Arena.ai compares user criticism signals of frontier coding agents

Based on 20,869 traces from Agent Arena: Code, Arena.ai compared user criticism signals of five top models from Anthropic and OpenAI against other agents. The analysis found that frontier models differ most noticeably in three dimensions: 6.9 percentage points fewer complaints about failure-fix loops and 1.9 percentage points fewer about ignoring project context; 6.4 percentage points fewer design issues and 5.8 percentage points fewer issues with producing usable results; and 5.6, 4.9, and 4.5 percentage points fewer complaints about instruction following, task completion, and breaking existing code, respectively. There are currently no follow-up developments.

View story:Arena.ai compares coding agent user criticism across 20,869 traces