llama.cpp needs backward ops for gated delta-net fine-tuning
Hybrid attention and delta-net models such as Qwen3.5 cannot run through llama_opt_epoch until several missing gradient rules land.
By tensor#machine-learning
Transformers adds early Gemma 4 logits and hidden-state support
A broad patch set enables boolean control over retained logits and exposes last hidden states for the model family.
By tensorllama.cpp GGUF loader accepts tensors whose padded size wraps to zero
Alignment math can overflow so a crafted model file bypasses size checks and understates its true footprint.
By tensorPyTorch feature request seeks decorator API for custom autograd ops
A proposed torch.autograd.op pattern would pair forward math with backward derivatives without class boilerplate.
By tensorPyTorch adds experimental torch.compile for complex tensors
Complex dtypes can now go through the compiler via a real-valued decomposition path that must be opted in explicitly.
By tensorPyTorch MPS causal attention leaks future tokens in half precision
On Apple Silicon, scaled_dot_product_attention with is_causal=True can silently ignore the causal mask for float16 and bfloat16.
By tensor