Matching stories Newest first
Hugging Face CEO says Netflix's OpenAI hack documentary is inaccurate
Clem Delangue, Hugging Face co-founder and featured subject of the Netflix film "AI Gone Rogue", says the documentary went "full doomer mode" and is not accurate.
Hugging Face model Spark-X2.6-4B, a compact on-device language model, is released
Spark-X2.6-4B keeps the Spark-X2.5-4B architecture and gains from improved post-training, with three thinking levels (none, low, high). The developer claims it matches or beats models around 9B parameters on open-source benchmarks.
Hugging Face blog shows learned memory tops the τ²-bench customer service leaderboard without a frontier model
IBM Research's ALTK-Evolve reportedly ranks first on τ²-bench customer service tasks using learned memory rather than a frontier model.
Ornith-1.5-122B, a Qwen-based finetune, released in GGUF and NVFP4 formats
Ornith AI introduces Ornith-1.5, which continuously generates its own training tasks and improves through reinforcement learning, extending Ornith-1.0 built on Qwen3.5 and Gemma4. GGUF and NVFP4 builds are on Hugging Face, with more details in the company blog.
ISTA-DASLab releases a Qwen3.8-27B prefiller model that streams from SSD for low-quant accuracy
The demo recovers accuracy for low-bit quantized Qwen3.8-27B while keeping VRAM usage low and decoding fast. It reportedly needs an Nvidia Blackwell GPU, and the poster has not tested it. For context longer than about 4k tokens, the poster expects no speed penalty.
Hugging Face policy chief says embedded AI evaluators will not be an easy fix
TechCrunch reports that Hugging Face's policy chief questions whether placing supposedly neutral third-party evaluators inside AI labs can solve the safety concerns raised by a rash of recent incidents.
Hugging Face publishes TorchNeuron guide for running Hugging Face models on AWS Trainium
The post covers running Hugging Face models on AWS Trainium chips through the TorchNeuron PyTorch backend, presented as a simpler path than before.