Loading…
Latest updates from the world of artificial intelligence
Latest updates from the world of artificial intelligence
Claude Opus 5 and GPT-5.6 Sol got the same three build briefs — a playable Pac-Man, an animated seaside fairground and a motorway interchange — one attempt each, with no browser to check their own work. All six builds run live in the article: play both games, watch the rest, and see which one holds up.
BitsMinds Lab
Read more →
We gave four Claude models — Fable 5.1, Opus 5, Fable 5 and Sonnet 5 — the same five build briefs at maximum reasoning effort, one attempt each, with no browser to check their work. All twenty builds run live inside the article. Fable 5.1 won four of the five rounds and finished only a single point clear, because the round it lost was the one where its code never ran at all.
BitsMinds Lab
Read more →
We gave Claude Opus 5, Fable 5 and Sonnet 5 the same three build briefs — an animated SVG fairground, a falling-sand physics sandbox and a self-solving Rubik’s cube — one run each, no retries, no browser. All nine builds run live inside the article. Opus 5 takes it 8–6–3, and the biggest surprise is the clock: the largest model was the fastest on every round, by a factor of nearly three.
BitsMinds Lab
Read more →An internal version of Astra resolved ten open problems across group theory, high-dimensional geometry, coding theory, quantum complexity, lattice cryptography and extremal combinatorics, OpenAI says — with the proofs formalised in Lean as machine-checkable certificates. That is a genuine methodological upgrade over May's Erdős result, which rested on expert sign-off. It also leaves the one question a proof checker cannot answer.
OpenAI’s GPT-6 Astra and Anthropic’s Claude Fable 5.1 got the same three build briefs — a motorway interchange, a seaside fairground and a cinematic rocket launch — one attempt each, at maximum reasoning effort, and neither could open a browser to check its own work. All six builds run live inside the article, faults and all: a bridge that hides a truck mid-crossing, a carousel that is not doing what it looks like it is doing, and two rockets with very different ideas of what a launch is. Score it yourself.
BitsMinds Lab
Read more →OpenAI's largest training run ever produced a model that drives a computer at superhuman speed and is the first rated Critical on cyber. Independent scoring puts its intelligence level with the model it replaces — at 2.5x the price.
Axios
Read more →Meta's fourth frontier model in five months takes the top spot on DeepSWE, near-saturates million-token retrieval, and undercuts every rival on price — as long as you let Meta train on your traffic.
The Register
Read more →Atlas is a single model that generates a minute of 1440p camera-controlled video, reconstructs scenes in 3D from a handful of photos, and outputs point clouds and Gaussian splats — beating specialist reconstruction models on their own benchmark.
World Labs
Read more →OpenAI says Astra is the first large language model to cross the Critical cybersecurity tier of its Preparedness Framework, scoring 100% on ExploitBench and discovering two zero-days on its own. Advanced cyber access goes to alpha testers first.
OpenAI
Read more →Anthropic’s Fable 5.1 doubles Terminal-Bench-Science, tops every published head-to-head against GPT-5.6 Sol, cuts cache reads 75% — and ships beside Mythos 5.1, a candidly riskier sibling for vetted defenders.
Anthropic
Z.ai’s new open-weight model scores 57.0 on the Artificial Analysis Intelligence Index — six points behind Claude Opus 5 and under two behind GPT-5.6 Sol, at $0.15 per million input tokens. It loses the flagship fight and wins the one that matters for volume: it beats Sonnet 5, Terra and Luna outright.
Artificial Analysis
Read more →The stealth model that appeared on OpenRouter under the provider name “stealth” has a name at last. Z.ai’s GLM-5.3-Flash is a 320-billion-parameter multimodal mixture-of-experts model with a million-token context window and MIT-licensed weights — reportedly, a preview served entirely on Chinese-made accelerators.
MarkTechPost
Read more →IBM published three open-weight models — 3B, 8B and 30B — with a switchable thinking mode across the family and reinforcement learning inside real software-engineering, terminal and web-search sandboxes for the two larger sizes. The 30B scores 57.0 on SWE-Bench Verified, and the whole family runs on hardware you own.
The Decoder
Read more →A reasoning model with a million-token context window appeared on OpenRouter on August 20 under the provider name "stealth". Five days later, no lab has admitted building it.
TechCrunch
Google shipped Gemini 3.7 Flash just three weeks after 3.6 Flash, claiming a jump from 34.4% to 43.6% on FrontierCode and from 49% to 65.3% on DeepSWE — and priced the introductory tier at half what 3.6 Flash cost.
Z.ai shipped GLM-5.3 on August 14 without retraining its base model. Every gain comes from scaled post-training — and on the company’s own coding benchmark it edges past Claude Opus 4.8 while spending roughly 60% fewer output tokens.
MarkTechPost
Read more →No blog post, no changelog, no press release — just an edit to the API pricing page. DeepSeek-V4-Pro-0813 is now generally available at $0.87 per million output tokens, with benchmark gains that nobody outside DeepSeek has reproduced.
OpenRouter
Read more →Artificial Analysis scored Grok 4.6 at 61 on its Intelligence Index — level with GPT-5.6 Sol at max, behind Fable 5 at 62 and Opus 5 at 63. It costs $2/$6 per million tokens against $5/$25 and $5/$30, and finishes long agentic tasks in ~53 turns where Opus 5 takes ~103. It also trails both rivals by eight points on the hardest software-engineering evals.
Artificial Analysis
Read more →OpenAI split its Daybreak security program into Blue and Red tiers and handed the Red tier a purpose-trained model that clears 95% of advanced cyber tasks — against 1.5% for the same frontier model with its guardrails intact.
The Decoder
Read more →Meta released Muse Glimmer under Apache 2.0 — a 30-billion-parameter agentic model distilled from Muse Spark that quantises under 20 GB and runs on a single consumer GPU.
Meta AI Research
Read more →OpenAI slowed work on its next frontier model after internal evaluations suggested Astra may cross the Critical cybersecurity threshold — a level no model has triggered before. Weights are locked down, some internal work is paused, and outside evaluators are being called in.
TechCrunch
Mistral's new open-weights safety classifier takes the moderation policy as a plain-language prompt at inference time — no retraining — and matches guard models seven times its size while running on a single 16 GB GPU.
Mistral AI
Read more →A hidden entry in Microsoft’s MAI Playground reveals MAI-Realtime, a full-duplex speech model that listens and speaks simultaneously across seventeen languages, with two configurable ways of deciding whose turn it is. Microsoft has confirmed nothing — but the component it would displace inside Azure’s Voice Live service is OpenAI’s GPT-Realtime.
TestingCatalog
Read more →The 2.4-trillion-parameter flagship went broadly available today through Alibaba Cloud's APIs and the new QwenWork beta, with open weights promised for next week. The parameter count is the least interesting part: this is Alibaba reversing course after keeping recent flagships closed. The performance claim behind it — second only to Claude Fable 5 — still has no independent benchmark, and would require a 14-point jump in one generation.
South China Morning Post
Read more →The Decoder