English translation is being prepared.
Matched fieldsTitle
←Search results
Search “Agents”
English translation is being prepared.
Matched fieldsTitle
English translation is being prepared.
Matched fieldsSummary
Figma VP of Engineering Josh Clemm shares the editing model's training method
On October 29, 2026, Figma VP of Engineering Josh Clemm publicly presented the training method for Figma's design editing model, explaining that it shifted from supervised fine-tuning (SFT) to reinforcement learning (RL) and post-trained an as-yet-unreleased editing model based on Qwen 3.8 27B. He noted that early training with Llama 3.3 70B on more than 250,000 samples was enough to compete with top models on editing tasks at the time. The Figma agent is now available to all users; in testing it is about 20% faster on typical tasks, outperforms or matches the previous production agent in 62% of expert comparisons, receives 33% higher user ratings, and results in 60% fewer undo operations.
orcement learning (RL) and post-trained an as-yet-unreleased editing model based on Qwen 3.8 27B. He noted that early training with Llama 3.3 70B on more than 250,000 samples was enough to compete with top models on editing tasks at the time. The Figma agent is now available to all users; in testing it is about 20% faster on typical tasks, outperforms or matches the previous production agent in 62% of expert comparisons, receives 33% higher user ratings, and results in 60% fewer undo operations.
Matched fieldsSummary
OpenAI launches GPT-6.1 Terra on ChatGPT Work and Codex
On October 27, 2026, OpenAI released and launched GPT-6.1 Terra, touted for its speed and efficiency, visual reasoning, and precise computer operation, with clearer and more natural writing as well, rolling it out that day to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex. On the same day, Arena announced that the model had entered Agent Arena and Code Arena; OpenRouter launched the model, saying it matches GPT-6 Astra on Agents' Last Exam at one-tenth the price per token; OpenClaw also announced that the model is now selectable. OpenAI co-founder Greg Brockman said Terra's computer use has crossed the threshold of a qualitative change and that using it is faster than operating a computer on his own. fal, meanwhile, used the fal Codex plugin to test turning Blender 3D content into Seedance 2.5 cinematic footage. On October 28, Arena announced that Terra ranked 14th in Code Arena: WebDev with a score of 1671, at a median cost of $0.40 per task, 28% lower than GPT-5.6 Terra. OpenAI's system card confirms that Terra is overall safer than older models, but its flag rates on items such as concealing uncertainty, circumventing restrictions, data exfiltration, and destructive operations
ore natural writing as well, rolling it out that day to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex. On the same day, Arena announced that the model had entered Agent Arena and Code Arena; OpenRouter launched the model, saying it matches GPT-6 Astra on Agents' Last Exam at one-tenth the price per token; OpenClaw also announced that the model is now selectable. OpenAI co-founder Greg Brockman said Terra's computer use has crossed the threshold of a qualitative change and that u
Matched fieldsSummary
Google DeepMind and Cambridge paper demonstrates AI agents autonomously discovering mathematical rules
Google DeepMind and Cambridge have published a paper demonstrating that a Gemini-based AI agent can autonomously discover new mathematics. The agent found and provided exact rules for 11 Busy Beaver Turing machines that had no prior analysis records. According to the paper, the agent first tested edge cases in a simulator, drafted informal proofs, and then completed the proofs in the Lean proof checker; all the relevant rules were verified through Lean. The paper states that this reflects AI moving from answering mathematical questions posed by humans toward open-ended discovery. Currently, this work is presented as a paper, and there have been no reports of further progress.
Matched fieldsSummary
Sarvam AI introduces a knowledge fabric architecture for enterprise agents
On October 21, 2026, Sarvam AI introduced a knowledge fabric architecture built for enterprise agents. The architecture consists of two layers, a knowledge tree and a domain ontology, compiling documents once into a structured layer that agents can navigate, query, and cite. According to its published results, on the Harvey LAB benchmark, answers using GLM 5.3 met 84.2% of expert grading criteria, higher than agentic RAG's 55.1%, with average task cost dropping from $0.64 to $0.28; on Sierra's tau-knowledge banking benchmark, it achieved accuracy comparable to the baseline using 64% fewer tokens. The knowledge tree is already used in the Sarvam cowork product, while the ontology is built domain by domain and enterprise client by enterprise client.
On October 21, 2026, Sarvam AI introduced a knowledge fabric architecture built for enterprise agents. The architecture consists of two layers, a knowledge tree and a domain ontology, compiling documents once into a structured layer that agents can navigate, query, and cite. According to its published results, on the Harvey LAB benchmark, answers using GLM 5.3 met 84.2% of expert grading criteria, higher than agentic RAG's 55.1%, with average task cost dropping from $0.64 to $0.28; on Sierra's t
Matched fieldsSummary
Chelsea Finn proposes optimizing information organization to extend LLM context
On October 22, 2026, Chelsea Finn reposted the new paper Librarian Agents by Anikait Singh et al., and proposed optimizing information organization to improve downstream utility. She pointed out that agentic search can scale to information of arbitrary size but tends to miss relevant content; whereas optimizing information organization for downstream use can bring 30+% improvements even in OOD domains. The tweet includes links to the paper and code. There are currently no subsequent development facts.
On October 22, 2026, Chelsea Finn reposted the new paper Librarian Agents by Anikait Singh et al., and proposed optimizing information organization to improve downstream utility. She pointed out that agentic search can scale to information of arbitrary size but tends to miss relevant content; whereas optimizing information organization for downstream use can bring 30+% improvements even in OOD domains. The tweet includes links to the paper and code. There are currently no subsequent development
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
Princeton and Google paper finds 4B model trained on agent states achieves higher task success rate
Princeton University and Google published a paper studying the performance of a 4B model trained on agent states in agent tasks. The paper reports that the most accurate decision model on isolated problems solved only 69.5% of agent tasks, while the 4B model trained on agent states solved 91.5%. Its harness delegates tool selection at each step to a fast decision model, adopting its answer only when confidence exceeds a threshold fitted on the agent's own runs, and trained AgentDecision-4B on the states the agent actually reached, including states after mistakes. On tasks similar to the training data, this type of 4B model reduced agent costs by 81.1% to 89.0% and increased speed by up to 2.5 times, with a success rate gap of within 3 percentage points; delegating all tool selections without a threshold lowered the success rate by 19.0 percentage points. The paper is available at arxiv.org/abs/2610.21412.
Princeton University and Google published a paper studying the performance of a 4B model trained on agent states in agent tasks. The paper reports that the most accurate decision model on isolated problems solved only 69.5% of agent tasks, while the 4B model trained on agent states solved 91.5%. Its harness delegates tool selection at each step to a fast decision model, adopting its answer only when confidence exceeds a threshold fitted on the agent's own runs, and trained AgentDecision-4B on th
Matched fieldsSummary
Anthropic releases report disclosing changes in the company's workforce structure
Anthropic released an internal research report analyzing AI's impact on the company's hiring and work organization, based on data including employment, AI usage, output metrics, employee surveys, calendars and recruiting records, as well as interviews with 17 employees. The report says the share of junior employees in technical departments fell from 21% in early 2025 to 14%, and the share of junior technical roles in hiring dropped from 27% to 9%; by August, interactive AI agents' working hours were six times those of employees, and per-engineer code commits reached five times the level of six months earlier. The report also notes that human attention has become the bottleneck for agent output, that employees are shifting toward meta-work of building monitoring and orchestration agents, and that hiring now places more emphasis on autonomy and judgment. Subsequent reporting says that 80% of the time spent in Claude sessions is agents waiting for human input, that in March agents' total working hours exceeded humans' for the first time, and that the share of interactive sessions employees use to manage agents rose from 10% to 22%.
f junior employees in technical departments fell from 21% in early 2025 to 14%, and the share of junior technical roles in hiring dropped from 27% to 9%; by August, interactive AI agents' working hours were six times those of employees, and per-engineer code commits reached five times the level of six months earlier. The report also notes that human attention has become the bottleneck for agent output, that employees are shifting toward meta-work of building monitoring and orchestration agents,
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
Nvidia releases Nemotron 3.5 Super VL and Bolt models
On October 15, 2026, Nvidia announced the launch of Nemotron 3.5 Super VL and the edge-oriented 3.5 Bolt models. According to Nvidia's official introduction, Super VL is a 120B-parameter open model aimed at always-on agent tasks, and the company says it has the highest throughput among comparable open models, 5 times faster than the next-best model; it also launched two edge models, Bolt 74B-A9B 24GB and Bolt E4B 4GB, with the former able to run agent tasks locally on RTX/DGX Spark and the latter aimed at voice and conversational agents on GeForce RTX and Jetson. Super VL and Bolt 74B-A9B are already available on Hugging Face, and Bolt E4B is scheduled for release on October 23. On the same day, SGLang announced Day-0 support for Super VL, reaching 243.9 tok/s on a single RTX 6000 PRO with the help of MTP; Prime Intellect and Nvidia used GRPO to post-train it for about 11 hours on 128 GB200 GPUs, raising its score on the public AutomationBench-AA set from 34.0% to 57.7%. That evening OpenRouter listed the model, and the next day Perplexity announced that Pro and Max subscribers and the Agent API can use it.
the launch of Nemotron 3.5 Super VL and the edge-oriented 3.5 Bolt models. According to Nvidia's official introduction, Super VL is a 120B-parameter open model aimed at always-on agent tasks, and the company says it has the highest throughput among comparable open models, 5 times faster than the next-best model; it also launched two edge models, Bolt 74B-A9B 24GB and Bolt E4B 4GB, with the former able to run agent tasks locally on RTX/DGX Spark and the latter aimed at voice and conversational a
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary