weights
12 stories and discussions about weights, aggregated from every source we track.
Exfiltrate LLM weights and data through GET requests
A short background on SageMaker real-time endpoints, then a measured comparison of Gemma 4 E2B's QAT w4a16 checkpoint against the full-size bf16 release on the same NVIDIA L4 endpoint: decode speed, parallel throughput, answers and cost.
A step by step deployment of Gemma 4 E2B with vLLM on a single Tesla T4 attached to a Compute Engine VM, and a measured comparison of the QAT w4a16 checkpoint against the bf16 reference on the same card.
Contribute to chrishayuk/larql development by creating an account on GitHub.
Attackers and defenders download the same LLM weights, but only the defender has to decide inline. Why that time budget creates the asymmetry, and where it does not hold.
The argument regarding LLM hysteria centers on the collapse of corporate claims under basic logical scrutiny, revealing their reliance on concept substitution, demagoguery, and the disregard of foundational legal…
We just released Ternary Bonsai 2 27B: 5.9 GB of weights for a 27B model. How much reasoning survived compression? We test it on a few evals that I found interesting and discuss where it retains the performance of the…
Upload files and keep a shareable link and SHA-256 checksum.
Automated blog of open conversational model releases on Hugging Face from labs with a model above the bar on the Artificial Analysis Intelligence Index.
A short background on SageMaker real-time endpoints, then a measured comparison of Gemma 4 E2B's QAT w4a16 checkpoint against the full-size bf16 release on the same NVIDIA L4 endpoint: decode speed, parallel throughput, answers and cost.