AllLLMRoboticsStartupsResearchEthicsBusinessHardwareAutonomousMedicineCreative

All news

10593
Axiom Math formalizes the "Enormous Theorem" in 43.1M lines of Lean in 24 days
LLM·TG: @ai_machinelearning_big_data (Machinelearning)

Axiom Math formalizes the "Enormous Theorem" in 43.1M lines of Lean in 24 days

Oct 28, 2026·3 min
Cline Pushes an Open Model to 11,840 Tokens per Second on a Single GPU
LLM

Cline Pushes an Open Model to 11,840 Tokens per Second on a Single GPU

Oct 28, 2026·3 min
Netflix speeds up Java AI agents: IntelliJ engine saves 20–30% on typical tasks, up to 65% at best
Business

Netflix speeds up Java AI agents: IntelliJ engine saves 20–30% on typical tasks, up to 65% at best

Oct 28, 2026·4 min
METR and Redwood Broke OpenAI's Agent Monitoring: 75–94% of Traffic Covered
Business

METR and Redwood Broke OpenAI's Agent Monitoring: 75–94% of Traffic Covered

Oct 28, 2026·4 min
ElevenLabs Launches Eleven Realtime: A Voice Agent in a Single Model, 98.8% on Big Bench Audio
Business·ElevenLabs Blog

ElevenLabs Launches Eleven Realtime: A Voice Agent in a Single Model, 98.8% on Big Bench Audio

ElevenLabs released Eleven Realtime v1 Mini and Max, speech-to-speech models that replace the ASR, LLM and TTS stack. Max responds in 0.57 s and answers 98.8% of…

Oct 28, 2026·3 min
14 AI models spot OLX scammers, but 37 of 42 quietly help with prepayment
Business·Хабр: Машинное обучение

14 AI models spot OLX scammers, but 37 of 42 quietly help with prepayment

A Kazakhstan developer tested 14 LLMs on 72 scam scenarios. Classification scores ranged from 0.93 to 1.00, but only 5 of 42 OLX prepayment replies included a…

Oct 28, 2026·3 min
Cloudflare Launches Auto Router: It Picks the Model So Employees Don't Have To, Saving Up to 30%
Business·Cloudflare AI Blog

Cloudflare Launches Auto Router: It Picks the Model So Employees Don't Have To, Saving Up to 30%

Cloudflare put Auto Router for AI Gateway into beta. In an internal test it cost $2.10 vs. $5.91 for Claude Opus 5.5, with an 86.6% task success rate vs. 96.6%.

Oct 28, 2026·4 min
T-Bank replaces a swarm of LLMs with a single 32B model and cuts query costs by 2.8–3.9x
Business·TG: @hiaimedia (Hi, AI • Новости технологий)

T-Bank replaces a swarm of LLMs with a single 32B model and cuts query costs by 2.8–3.9x

At EMNLP 2026, T-Bank researchers showed how one Qwen3-32B-based model covers 200+ services and beats Qwen3-235B on corporate benchmarks: 69.6 vs 65.8.

Oct 28, 2026·3 min
Top themes
01Cloudflare releases Clairic: a data extraction model with no output tokensOct 27, 2026·3 min02T-one and GigaAM cascade: a draft of the subtitles during the call, a clean transcript afterOct 27, 2026·3 min03OpenAI details how its FDE team works: 90% of custom code gave way to Codex skillsOct 27, 2026·3 min04Anthropic finds experts follow an AI agent's lead in 86% of contested choicesOct 27, 2026·3 min05Anthropic reveals how it reviews code that Claude writes 90% ofOct 27, 2026·3 min
Open-source Hob brings Claude Cowork-style workflows to your laptop, no cloud required
LLM·Hacker News AI

Open-source Hob brings Claude Cowork-style workflows to your laptop, no cloud required

Oct 27, 2026·3 min
Anthropic files S-1: half-year revenue of $16.3 billion, with a valuation of up to $2 trillion expected
Business·TG: @data_secrets (Data Secrets)

Anthropic files S-1: half-year revenue of $16.3 billion, with a valuation of up to $2 trillion expected

Oct 27, 2026·2 min
Anthropic Urges Teams to Fold Subagents Back In: A Single Loop With Skills Beats Orchestrators
Business·TG: @ai_product (AI Product / Igor Akimov)

Anthropic Urges Teams to Fold Subagents Back In: A Single Loop With Skills Beats Orchestrators

Oct 27, 2026·3 min
A ~/.claude Architecture With 200+ Skills: Keeping the System Prompt Under 50K Tokens and Paying $0.15 Per Session
Opinion·Жемал Хамидун · Авторская колонка

A ~/.claude Architecture With 200+ Skills: Keeping the System Prompt Under 50K Tokens and Paying $0.15 Per Session

Oct 27, 2026·3 min
Cloudflare rebuilds Containers for agents: startup is 6x faster
Business·Cloudflare AI Blog

Cloudflare rebuilds Containers for agents: startup is 6x faster

Cloudflare rewrote Containers for AI agents: code picks the image and resources, median startup fell from 4.05 s to 648 ms, and 100,000 containers launch in 5.4…

Oct 27, 2026·4 min
ElevenLabs Hands Its Support Desk to an AI Agent: 22,000 Tickets a Month, 65% Without Humans
Business·ElevenLabs Blog

ElevenLabs Hands Its Support Desk to an AI Agent: 22,000 Tickets a Month, 65% Without Humans

ElevenLabs runs its own support on ElevenAgents: the agent fully resolves 65% of requests, replies in 60 seconds, and 70% of its conversations get positive ratings.

Oct 27, 2026·3 min
Anthropic Advises Cutting Half Your Subagents: One Loop Ran Twice as Fast
Business·TG: @cgevent (Метаверсище и ИИще)

Anthropic Advises Cutting Half Your Subagents: One Loop Ran Twice as Fast

Anthropic's Applied AI team reviewed enterprise deployments: a single agent with skills often beats a multi-agent harness on quality, cost and speed. One client saw a ~2x gain.

Oct 27, 2026·3 min
OpenAI Turns GPT-6 Astra Loose on Legacy Code: 800 Thousand Lines Deleted in Two Weeks
Business·TG: @ai_product (AI Product / Igor Akimov)

OpenAI Turns GPT-6 Astra Loose on Legacy Code: 800 Thousand Lines Deleted in Two Weeks

Oct 27, 2026·3 min
Cloudflare cuts the cost of AI code review by 68%, from $2.43 to $0.77
Business·Cloudflare AI Blog

Cloudflare cuts the cost of AI code review by 68%, from $2.43 to $0.77

Oct 27, 2026·5 min
Cloudflare launches beta of a gateway where AI agents pay via HTTP 402
Business·Cloudflare AI Blog

Cloudflare launches beta of a gateway where AI agents pay via HTTP 402

Oct 27, 2026·3 min
Berkeley Describes Adaptive Parallel Reasoning: The Model Decides When to Branch
Research·Berkeley BAIR

Berkeley Describes Adaptive Parallel Reasoning: The Model Decides When to Branch

Oct 27, 2026·3 min
Meta opens Muse Spark at 816B, Reflection releases Beam under Apache 2.0
Business·Latent Space (swyx)

Meta opens Muse Spark at 816B, Reflection releases Beam under Apache 2.0

Oct 27, 2026·4 min
NTU Scientists Teach AI Model to Skip Video: 59% Fewer Tokens
Research

NTU Scientists Teach AI Model to Skip Video: 59% Fewer Tokens

Oct 27, 2026·3 min
Nepali Text "Weighed" Three Times More Than English — Tokenizer Fix Couldn't Fully Solve It
Research

Nepali Text "Weighed" Three Times More Than English — Tokenizer Fix Couldn't Fully Solve It

Oct 27, 2026·3 min
SAP researchers teach AI agents to stop flip-flopping between different solutions to the same task
Research

SAP researchers teach AI agents to stop flip-flopping between different solutions to the same task

Oct 27, 2026·3 min
Microsoft Research Unveils Magnetar: a zkVM 10–39x Faster Than Alternatives
Business·Microsoft Research

Microsoft Research Unveils Magnetar: a zkVM 10–39x Faster Than Alternatives

Microsoft Research releases Magnetar: proofs of Rust, Go and C program execution, 10–39x faster than rivals, with 12 ms verification and security that rests only on…

Oct 27, 2026·4 min
AWS teaches voice AI agents to start speaking from the first sentence with Kokoro-82M
LLM·AWS Machine Learning Blog

AWS teaches voice AI agents to start speaking from the first sentence with Kokoro-82M

AWS showed how to stream speech from an AI agent using Kokoro-82M on SageMaker: synthesis starts with the first sentence instead of after a five-second wait for the…

Oct 27, 2026·4 min