Search “Agents”

a, focused on completing complex computer operation tasks at lower cost, with particular strength in visual reasoning and desktop software operation. At medium reasoning effort on Agents' Last Exam, its score is 15 percentage points higher than GPT-5.6 Terra, at about one-third the cost of the latter; at the highest reasoning effort, the cost per task is about one-ninth that of Claude Opus 5.5. The model is now available to Plus, Pro, Enterprise, and Edu users, can be used in ChatGPT Work and Co

Matched fieldsSummary、Body

g capabilities. The standard API price is $1 per million input tokens and $5 per million output tokens, 50% lower than GPT-5.6 Terra and GPT-6.1 Sol. At medium reasoning effort on Agents' Last Exam, its computer operation score is 15 percentage points higher than GPT-5.6 Terra, performing on par with GPT-6 Astra while costing only one-tenth as much per token. The model is now available to Plus, Pro, Enterprise, and Edu users of ChatGPT Work and Codex, but GPT-6.1 Sol and GPT-6 Astra are still re

Matched fieldsSummary、Body

.6 Terra, with API pricing at $1 per million input tokens and $5 for output, half that of GPT-6.1 Sol. The model supports a 1.05M token context and a maximum output of 128K, targets visual reasoning, computer operation and writing, and is offered only through OpenAI-hosted API, ChatGPT Work and Codex. On Agents' Last Exam v1 it scores 60.0% at the highest effort tier at $1.01 per task, and 85.5% on MMMU-Pro vision, but only 50.3% on Terminal-Bench 4.0, trailing GPT-6.1 Sol and Claude Sonnet 5.5.

Matched fieldsSummary、Body

visual understanding and computer operation capabilities, with the standard API price reduced by 50% compared to both GPT-5.6 Terra and GPT-6.1 Sol. At medium reasoning effort on Agents' Last Exam, its score is 15 percentage points higher than GPT-5.6 Terra, at about one-third the cost; at the highest reasoning effort on Terminal-Bench 4.0, it completes 50.3% of tasks. The model is available starting today to Plus, Pro, Enterprise, and Edu users of ChatGPT Work and Codex, with the API model nam

Matched fieldsSummary、Body

, visual reasoning, coding, and everyday office work. API pricing is $1 per million input tokens and $5 per million output tokens, 50% lower than GPT-5.6 Terra and GPT-6.1 Sol. On Agents' Last Exam, at medium reasoning effort it scores 15 percentage points higher than GPT-5.6 Terra at about one-third the cost, and completes tasks more than 3x faster than GPT-6 Astra and GPT-6.1 Sol. Internal benchmarks show the error rate at low reasoning effort dropped from 53.6% to 16.7%, and it scores 50.3% o

Matched fieldsSummary

Matched fieldsSummary、Body

ng. Its standard API price is 50% lower than that of GPT-5.6 Terra, at $1 per million input tokens and $5 per million output tokens. OpenAI says that at medium reasoning effort on Agents' Last Exam, it improves on GPT-5.6 Terra by 15 percentage points, and at the highest reasoning effort its token efficiency is more than three times that of Claude Opus 5.5, while its price per token is one quarter of the latter's. Starting October 26, the model is available to Plus, Pro, Enterprise, and Edu user

Matched fieldsSummary、Body

Matched fieldsSummary、Body

cached input reduced to $0.05. Artificial Analysis measured that at maximum effort, the model's intelligence index is 6 points higher than GPT-5.6 Terra, with per-task cost of $0.40 versus $1.40, and gains of 7 to 13 points at the low effort tier. In AA-Omniscience, the model's guessing rate on questions where it does not know the answer is 38%, down from 88%, while accuracy improved by 2 points; it scored 48 on the Coding Agent Index v2.0, and improved by about 20 points on Terminal-Bench 4.0.

Matched fieldsSummary、Body

AWS released a solution that uses AWS CDK to deploy enterprise-grade agentic retrieval on a managed Amazon Bedrock Knowledge Base: the reasoning agent invokes two knowledge bases as MCP tools through AgentCore Gateway, and AgenticRetrieveStream performs multi-turn planning, internal reranking, and returns answers with citations. The managed knowledge base (Type: MANAGED) takes over ingestion, storage, indexing, and retrieval, eliminating the need to build your own OpenSearch Serverless collectio

Matched fieldsSummary、Body

the Artificial Analysis Intelligence Index, at the highest effort level it scores 6 points higher than GPT-5.6 Terra, with a cost per task of $0.40 versus $1.40; at the low and medium effort levels it sits at the frontier of intelligence and cost. On the AA-Omniscience benchmark, the model's guess rate when it does not know the answer dropped from 88% to 38%, while accuracy simultaneously improved by 2 points; Terminal-Bench 4.0 improved by about 20 points, and Coding Agent Index v2.0 scored 48.

Matched fieldsSummary、Body

The author shared the process of using Codex and Claude to reverse engineer and retrofit real-world devices: first enable ADB debugging on the TV, then have the Agent disable system packages, clear boot ads, and rebuild a clean launcher, and replicate the same method on three other TVs. The team also used Claude to reverse engineer 4 fill lights and built a unified web console, and used Codex to build a multi-camera console for a Sony camera and a DJI Pocket 3 and deploy it to Feishu. The author

Matched fieldsSummary

miting and unnecessary quota consumption of the K3.1 Pro subscription plan, stating that this was its own problem and should have been fixed earlier. All subscribers receive a one-time usage reset, restoring 25% of the monthly quota (reset to zero if usage is below 25%), equivalent to one week's usage, and the usage bar will be updated before the end of the day PST. Kimi has also adjusted the plan's parameters to reduce unnecessary subagent calls and output length, making the quota more durable.

Matched fieldsSummary、Body

Harrison Chase and Caspar Broekhuizen of LangChain published an article summarizing the five core components shared by its three internal agents for marketing, sales, and engineering: agents live in existing workplaces such as Slack, a distinction between acting on behalf of users and a fixed identity, global and user-level memory, code execution within a sandbox, and a large number of runs triggered by schedules or events. The article uses GTM Agent, OpenSWE, and Content Studio as examples to e

Matched fieldsSummary、Body

Matched fieldsSummary、Body

d it provides builds for macOS, Windows, and Linux. Its DOCX editing approach archives the original file by hash, marks only the paragraphs that have become dirty, and generates OOXML fragments to splice back into document.xml, reducing formatting damage to unmodified content. The project splits agent-core, the AI provider abstraction, and the format engine into shared packages, but AI calls require logging into a Genspark account and go through its server; model API Keys are not stored locally.

Matched fieldsSummary

Google DeepMind had four Gemini 4 Argon agents team up to play Minecraft from scratch, defeating the Ender Dragon after 27 hours of real time, during which they died 605 times, wrote more than 2,700 programs, and produced a total of 63,300 lines of code. The four agents had no human-written programs to rely on; chopping trees, mining, and crafting equipment all required them to write and test their own code, and they also shared experience with each other and reported the locations where teammat

Matched fieldsSummary、Body

for product iteration, core component R&D, and algorithm platform development. The company was founded by Zhao Changyi, a PhD from Jilin University, and has launched the FaceSoul Agent full-stack emotional interaction solution, which uses a cable-driven approach to achieve 32 facial degrees of freedom and features a self-developed lightweight on-device expression generation model. The company has launched four major series—Lingshu M1, Lingshu W1, Lingyun, and Lingneng—and within four months of

Matched fieldsSummary、Body

Cognition has released the open-source, free Devin Vulnerability Fixes plugin, which includes three Agent skills—fix-php-vuln, fix-node-vuln, and fix-gitlab-vuln—under the MIT license, targeting PHP2Shell, the Node.js October security release, and the self-managed GitLab vulnerability CVE-2026-110264 with a CVSS score of 10.0. The skills execute in five steps: read-only inventory, approval, target version verification, before-and-after testing, and one PR per repository. They can run in agents t

Matched fieldsSummary、Body

. Frontier model training is no longer allowed to access the real internet; instead, models practice browsing and search in an offline web environment. In red-team experiments, an Agent, under conditions prohibiting network communication and with no shared writable files, used side channels such as CPU caches to exchange information; in another test, an internal research model exploited a DNS filtering gap to contact an external chatbot. The system raised an alert within 15 minutes, but the auto

Matched fieldsSummary、Body