SevenTnewS

Lab

Research, experimentation and open source: hardware, IoT, robotics, biotech and edge AI.

12 published articles

Featured3 min read

Market momentum

Nvidia just crossed $3 trillion again. The AI spending machine isn't slowing down.

Nvidia's market cap crossed $3 trillion as AI infrastructure demand keeps growing. The milestone underscores that enterprise GPU purchases remain strong despite growing competition from AMD and startups offering cheaper inference chips.

2026-07-16

5 min read

Machine Learning Theory

Google just proved why diffusion models invent, not just copy

Google researchers reveal that the creativity of diffusion models stems from a 'score smoothing' effect caused by neural network regularization. This theoretical framework explains why models interpolate between training data points rather than merely memorizing them, opening the path for controlled novelty in generative AI.

2026-07-16

4 min read

AI Hardware

Alibaba's bet against the touchscreen is smarter than it looks

Alibaba's Qwen Glasses strategy reveals a broader industry shift: AI glasses are approaching a convergence of capability, user readiness, and viable economics. The winners will be those that build complete ecosystems, not just clever devices.

2026-07-13

3 min read

Agent evaluation

Sandbox benchmarks are hiding how agents really fail, HKU just built the fix

UniClawBench evaluates proactive agents across five fundamental capabilities in 400 bilingual real-world tasks, using live Docker containers and a three-agent closed-loop evaluation. It disentangles base model abilities from framework choices, revealing where agents truly break.

2026-07-13

Featured3 min read

Robotics Research

Robots that adapt without retraining? This new framework might actually deliver

A new framework called In-Context World Modeling (ICWM) allows robot policies to adapt to novel setups, like different camera angles or robot bodies, without retraining. By treating system identification as an in-context problem, ICWM uses task-agnostic interactions to infer world dynamics before task execution, outperforming standard VLA baselines in simulations and real-world tests.

2026-07-11

4 min read

AI Infrastructure

The next trillion-dollar bottleneck in AI isn't algorithms, it's power and cooling

The explosive growth of generative AI is exposing a hard limit: the physical infrastructure needed to train and serve models at scale. Hyperscalers, chip designers, and a wave of energy-focused startups are pouring capital into rethinking data centers, cooling, and power delivery, betting that the next frontier of AI competition is infrastructural, not algorithmic.

2026-07-10

Featured8 min read

Smart Lock Review

The Schlage Sense Pro locked my front door before I could unlock it, then I fixed it

The Schlage Sense Pro uses ultra-wideband to unlock your front door as you approach, no keys, no codes, no phone taps. It is the first hands-free system this reviewer trusts, but its $399 price tag and exclusive reliance on Apple Home Key make it a hard sell for Android users, at least for now.

2026-07-09

Featured4 min read

Robotics AI

A compact robot model just beat multi-sensor systems with one camera and no depth sensors

Mistral AI launches Robostral Navigate, a compact 8B model that lets robots navigate complex indoor environments using just one camera. It beats multi-sensor approaches by 4.5 points on the R2R-CE benchmark, runs on wheeled, legged and flying robots, and was trained efficiently with prefix-caching to slash token count 22-fold.

2026-07-08

Featured4 min read

Data Engineering

JPEG at quality 92: the boring data engineering choice that made Photoroom's 7B model work

Photoroom releases the fourth part of its PRX model development series, detailing a data pipeline that prioritizes breadth and consistency over per-image perfection. The post reveals that JPEG compression at quality 92 is indistinguishable from PNG for training, and that the fastest captioner won over the most accurate one when scaling to hundreds of millions of images.

2026-07-06

5 min read

Genomic Medicine

Talos shows automated genome reanalysis can deliver new diagnoses in weeks, not years

Talos reframes genome reanalysis as a continuous, automated program rather than a rare manual event. Across a prospective cohort of 4,735 undiagnosed patients, it yielded 241 new diagnoses (5.1% additional yield) within weeks of new evidence being published, while requiring analysts to review only one new variant per 200 patients per month.

2026-07-05

5 min read

Benchmarks & Tests

Treble Technologies and Hugging Face Launch FFASR Leaderboard for Far-Field Speech Recognition

The new FFASR Leaderboard from Treble Technologies and Hugging Face evaluates ASR models across nine conditions including reverberation, background noise, and microphone distance. Early submissions show far-field WER at low SNR is several times higher than near-field performance, highlighting the need for acoustically robust models.

2026-07-02

Featured3 min read

Performance

Gemma 4 runs 90% faster in Ollama 0.31 with a trick that needs no config

Ollama 0.31 introduces multi-token prediction for Gemma 4 on Apple Silicon, achieving near 90% faster token generation on coding benchmarks. The speedup comes from an auto-tuned draft model and a custom MLX kernel that eliminates redundant weight reads.

2026-06-29