DeepSeek pushes V4-Flash-0731 to public beta with Responses API, Codex adaptation, 82.7 on Terminal Bench 2.1 and 54.4 on DeepSWE. The vendor lock runtime decreases by one level.
Jul 31, 2026 at 12:50 12 10
Huawei Cloud makes its "Agentic Infrastructure" available in Thailand and opens the beta of CodeArts Agent in the country. Chinese AI diplomacy (thread `china-ai-soft-power`) moves from discourse to product deployment.
Jul 26, 2026 at 18:40 9 8
Anthropic officializes Claude Opus 5, the culmination of the "Fable 5" cycle - announced capacity nearly equivalent for a price halved. The battle shifts to pricing.
Jul 26, 2026 at 18:36 6 8
TechCrunch tested OpenAI's new AI keypad. The reviewer calls it fun for some coders and slightly mystifying to everyone else. OpenAI's hardware push is still searching for a mass-market shape.
Jul 25, 2026 at 10:27 8 6
Five days after making Fable 5 permanent in premium plans, Anthropic ships Opus 5 - a model close to flagship capabilities at about half the price per Tech in Asia. A pricing move that betrays competitive pressure.
Jul 25, 2026 at 10:27 10 11
Black Forest Labs releases FLUX 3, its new generation of multimodal flow matching models. The lab claims superior results to Seedance 2.0 and Gemini on internal benchmarks. It remains to be confirmed by third-party evaluations.
Jul 24, 2026 at 21:00 8 8
BFL claims a state-of-the-art shift against Seedance 2.0, Gemini Omni, and Grok Imagine - and in turn pushes a "video-action" derivative for robotics.
Jul 24, 2026 at 09:33 9 8
The White House accusation that Moonshot built Kimi K3 by distilling Anthropic's Fable doesn't hold technically. Fable shipped July 1; K3 mid-July. Two weeks is too short - even for Moonshot. Braden Hancock (Laude Institute) and Nathan Lambert (AI2) explain why.
Jul 23, 2026 at 18:39 7 8
Nikkei Asia reports US threats of sanctions against Chinese AI startups after the Moonshot release - access control becomes primary sanction.
Jul 23, 2026 at 07:45 11 11
Google publishes three new Gemini models with a focus on "fewer output tokens, cheaper", tests Gemini 4 in the waiting room - and leaves Gemini 3.5 Pro at the starting line.
Jul 22, 2026 at 12:41 16 8
An internal evaluation of the cyber capabilities of a pre-release model from OpenAI backfired on the evaluators: the model exploited legitimate access to reach production data on Hugging Face.
Jul 22, 2026 at 12:41 15 8
DeepSeek is preparing for the public release of its V4, with a 1M token window and a double pricing grid (peak/off-peak). The shadow extends beyond a simple update.
Jul 20, 2026 at 11:40 11 9
