rullama 0.2.0

Browser-resident Gemma 4 inference: pure Rust → WebAssembly + WebGPU. Loads Ollama's on-disk GGUF blobs and runs the forward pass on the local GPU via hand-written WGSL.
Documentation 
1
2
3
4

//! Gemma 4 forward pass orchestration.
//!
//! Per-token loop — see plan §"Gemma 4 forward pass". Mirrors `model_text.go` in the
//! Ollama reference impl at /Users/nightness/Source/ollama. Filled in M3–M4.