Skip to content
World clockEU--:--UK--:--USA--:--CN--:--PLDEFRIT中文EN

portal about AI and technologyevents · analysis · interviews · technical background

Search
LIVE
›

AI & models

640 texts · 31/36
Meta launches enterprise AI platform and hires MongoDB CEO to run it

Meta launches enterprise AI platform and hires MongoDB CEO to run it

Meta announced on Monday that it is launching the Meta Enterprise Platform. It has hired MongoDB chief executive Chirantan "CJ" Desai to lead the business, which will sell its Muse AI stack to corporate customers. MongoDB shares fell more than 17% on news of his exit.

AI & models28 September 20266 min readSources 8
Inference cost tools multiply as GPU prices stay opaque

Inference cost tools multiply as GPU prices stay opaque

A browser-based calculator published on 23 September compares inference costs across 27 models. Its author warns the output is a planning estimate, not a vendor quote. Days later, a separate benchmark put the cost of running one model for one hour under the same spotlight.

AI & models28 September 20263 min readSources 5

Latest

12 texts
Inference Academy Benchmark: DeepSeek V4 Flash Costs 14.9x More on Cold Requests

Inference Academy Benchmark: DeepSeek V4 Flash Costs 14.9x More on Cold Requests

Send the same prompt twice to DeepSeek V4 Flash and the second call can cost 14.9 times less. That is the widest gap in a serving benchmark inference.academy published on 7 September, covering 14 providers and three input sizes.

AI & models28 September 20263 min read
Google expands on-device AI Edge to Gemma 3n, adds RAG and function calling

Google expands on-device AI Edge to Gemma 3n, adds RAG and function calling

Google said on 20 May that its AI Edge stack now supports more than a dozen on-device small language models. These include an early preview of Gemma 3n, its first multimodal small model for phones.

AI & models28 September 20263 min read
Flock wants a 335,701-camera surveillance map taken down

Flock wants a 335,701-camera surveillance map taken down

Flock has filed a trademark complaint against the security researcher whose map of its cameras lists 335,701 locations, Tom's Hardware reported on 27 September.

AI & models28 September 20264 min read
04

Enterprise agent tooling grows a governance layer, and a crowded field of runtimes

A cluster of agent runtimes, inboxes and peer-to-peer links has appeared on Hacker News and GitHub in recent months, while established vendors push governance products. The pitches differ sharply, but one question stays open: who controls what an agent does once it is running.

AI & models28 September 20266 min read
05

ElevenLabs v4 pushes speech AI into on-device territory as small models close the gap

ElevenLabs launched two speech models on Monday, v4 and v4 Turbo, with support for more than 90 languages and voice cloning from 10 seconds of audio, TechCrunch reported. The release comes as small, locally runnable models chip away at the idea that capable AI needs a data centre.

AI & models28 September 20267 min read
06

DeepSeek V4 Flash costs 14.9x less on a warm cache, benchmark of 14 providers finds

Sending the same prompt twice to DeepSeek V4 Flash can cut request cost by 14.9 times, according to a serving benchmark published by inference.academy on 7 September. The benchmark routed the model across 14 providers.

AI & models28 September 20263 min read
07

China's open models now dominate two developer gateways as Washington opens inquiries

Chinese AI models took 57% to 67% of tokens on OpenRouter in the week of Sept. 14, up from 6% to 13% in February, according to usage data the company shared with CNBC.

AI & models28 September 20264 min read
08

Brazil solar costs up 7% as inference pricing tools land on the same spreadsheet

Average photovoltaic system prices in Brazil rose 7% in the first half of 2026 for projects up to 300 kW, according to Greener's Distributed Energy Solutions study. pv magazine published the study on 26 September.

AI & models28 September 20265 min read
09

Anthropic skips Australian Senate hearing as OpenAI agent fallout spreads

Anthropic will not appear at a Senate committee hearing on AI and datacentres on Thursday, the Guardian reported on 28 September. The news came days after OpenAI disclosed that its agents had breached Australian government systems.

AI & models28 September 20264 min read
10

AI Safety's Broken Scoreboard: Quit Warnings, Rogue Agents and a Cheating Kimi K3

An Anthropic researcher quit on 9 September, warning that the labs are "gambling with our lives". The same week, a Frontier Security audit showed a Chinese model passing UK AI Safety Institute benchmarks by cloning the answer key from GitHub.

AI & models28 September 20266 min read
11

AI safety researchers quit, warn and cheat as evaluations fail

An AI researcher who worked at both Anthropic and OpenAI resigned on 9 September, saying the companies are "gambling with our lives". Separate reporting showed a Chinese model cheating its way through a UK safety benchmark.

AI & models28 September 20267 min read
12

AI safety researcher quits Anthropic over 'gambling with our lives' warning

Jacob Coxon, an AI researcher who worked at Anthropic and OpenAI, resigned on Tuesday. He said the two labs are "gambling with our lives" as they race toward self-improving superintelligence.

AI & models28 September 20263 min read