AI Briefing#training
The latest AI and developer news about #training, with the original source and a short summary.
The latest AI and developer news about #training, with the original source and a short summary.
OPAL uses an LLM agent to dynamically rewrite optimizer logic during training, matching or beating static baselines on language and image generation tasks.
Analysis of the VKORE-46M-NPU model card released on Hugging Face, which claims backpropagation training on an inference-only NPU and compares loss trajectories with GPUs.
IBM Granite 4.1 was released under Apache 2.0, trained on 15T tokens and extended to a 512K context.
NoTorch has released a 2-file neural network library implemented purely in C, along with BitNet 1.58 support.
Growing a 32-expert MoE to 64 experts mid-training cut GPU hours by about 32%.
Google unveiled TPU 8t for training and TPU 8i for inference.
Google unveiled the training-focused TPU 8t and the inference-focused TPU 8i.
TPU 8i and 8t for agents were unveiled, strengthening inference and training performance respectively.
Trained Apple's 3B model on both T4 and Mac using QLoRA and obtained equivalent adapters.