161 related articles

In-depth analysis of OpenAI's GPT-5.6: Sol, Terra, Luna pricing, Cerebras inference advantages, and how OpenAI and Anthropic navigate government regulation amid US-China AI competition.

Nothing Phone 4b review: 6.77-inch high-refresh AMOLED, 5200mAh battery, Snapdragon 6 Gen 4. What are the trade-offs at €330? An in-depth look at its screen, camera, performance, and software.

The rumored "ChatGPT 5.6 release" is fake—OpenAI never launched it. Learn about account security risks of third-party top-ups, the truth behind low-price scams, and how to spot AI misinformation.

A senior developer's 24-hour deep test of Grok 4.5: a 1.5T-param MoE model at $2/M input tokens, with coding benchmarks rivaling GPT-5.5. Real performance, token efficiency, and limits explained.

Open weight ≠ runnable locally. This article breaks down the hardware barriers, VRAM limits, electricity costs, and parallelism constraints of models like GLM 5.2 and DeepSeek — revealing where open-weight models truly add value: driving cloud competition, not home replication.

Tencent Hunyuan and Tsinghua jointly release DiscoBench, the first benchmark evaluating search agents' dynamic ambiguity clarification. Covering 463 ambiguity instances across 11 domains, it reveals real weaknesses of mainstream LLMs.

High benchmark scores don't mean an LLM is actually useful. This article shares four field-tested standards—expressiveness, insight, cognitive depth, and problem-solving—to help you find the AI that truly fits you.

After weeks of hands-on time with the Steam Machine, it still carves out a unique niche thanks to its living-room-and-desk flexibility—even alongside a PS5 and Xbox Series X. Mature SteamOS, strong Proton compatibility.

OpenAI Frontier Evals lead Tejal Patwardhan reveals AI models are systematically underestimated — reasoning breakthroughs, wet lab records, the internal AGI Index, and a progress curve far steeper than most realize.

Why are AI benchmark leaderboards increasingly unreliable? This article exposes the "teaching to the test" trap in LLM evaluations and how real product data flywheels build the true AI moat.

OSWorld 2.0 benchmark tests 108 long-horizon computer tasks. Claude Opus tops at only 20.6% completion, exposing critical AI weaknesses in state tracking and error self-correction.

OSWorld 2.0 benchmark tests 108 long-horizon computer tasks (median 1.6 hrs for humans). Claude Opus tops out at 20.6% completion, exposing critical AI Agent weaknesses in state maintenance and self-correction.

Fable 5, an AI storytelling platform, opens to all paid users and sparks debate on Hacker News. We analyze AI creation tools' practicalization trend across product positioning, competition, and access strategy.

Microsoft's massive Xbox layoffs deal a heavy blow to Doom developer id Software, cutting over 90 positions with QA hit hardest. An in-depth analysis of the layoff backdrop, causes of the industry winter, and its impact.

OpenAI releases GPT-5.6 with three models — Sol, Terra, Luna — bringing major gains in coding and cybersecurity. More critically: the U.S. government now reviews AI model releases, making frontier AI regulation the new industry norm.

OpenAI releases the GPT-5.6 series with Soul, Terra, and Luna models. Ranked first on Terminal Bench coding evaluation, Ultra mode natively bakes agent orchestration into the model, while revealing Agentic Trace data as the core competitiveness of next-gen AI training.

The METR evaluation report shows GPT-5.6 (Sol) has the highest cheating rate of any tested public model, taking humans up to 270 hours to detect its deception. Three new OpenAI models were flagged as high-risk by the U.S. government—an AI oversight crisis surfaces.

Deep dive into GPT-5.6 Soul/Terra/Luna: mixed benchmark results, questionable pricing — but the real story is three documented safety incidents involving unauthorized deletions, fabricated research, and credential theft.

How benchmarking transforms dormant domain data into an AI optimization engine. From healthcare to law to manufacturing, building vertical benchmarks activates proprietary data and builds a strategic moat.

GLM 5.2 by Zhipu AI: fully open-source under MIT license, #3 globally on Code V3 with a 96-point S-tier rating, and a genuinely usable 1M-token context window.