Search “Agents”

Arena.ai says Gemini 4 Carbon ranks second on Agent Arena

On October 28, 2026, Google DeepMind released Gemini 4 Carbon, calling it its strongest model, ranking first with 59 points on the Artificial Analysis Intelligence Index v5.0, and scoring 88.2% on SEC-bench Pro, 54.8% on ExploitGym, and 71.0% on CWE-bench. The model targets long-horizon engineering tasks and was first opened to some cybersecurity defenders, then expanded through the Fairwind Program, and afterward made available to developers, enterprises, and consumers. On October 29, Arena.ai said Gemini 4 Carbon ranked second in Agent Arena with a net improvement of 16.2% and pushed the Pareto frontier forward; the same day, Yuchen Jin cited Artificial Analysis data to raise reservations, noting that the evaluation used the Antigravity CLI. On October 30, the Artificial Analysis Intelligence Index showed it in first place with 59 points, 1 point higher than Claude Opus 5.5. On October 31, Perplexity CEO Aravind Srinivas shared an early evaluation, saying it set a new SoTA on the WANDR benchmark.

View story:Google releases the Gemini 4 Carbon model

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

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.

Arena.ai Publishes Per-Task Cost Comparison of Models on Agent Arena

Arena.ai published a per-task cost comparison of models on Agent Arena on October 30, 2026. Statistics show that OpenAI's GPT-6.1 Sol (Max) costs $1.91 per task, 85% lower than Claude Opus 5.5 (Max) at $12.51 and 87% lower than GPT-6 Astra (Max) at $14.93. Arena.ai noted that these two cost advantages come from different sources: relative to Astra, Sol's advantage mainly comes from a lower per-token price; relative to Opus, it mainly comes from consuming fewer tokens per task. Per-task cost is calculated by multiplying the number of input, output, and cache-read tokens by their respective unit prices, and Arena.ai identifies the reasons for the cost reduction by comparing prices and token usage.

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