🦀 Decoder-only LLM built from scratch in pure Rust using Candle — no Python, no PyTorch. Gated DeltaNet + sparse attention, fine-grained MoE, native video/document understanding, long-horizon tool agents, quantization-aware training. Scales: Tiny (25M) to Large (1.3B).
rust machine-learning deep-learning lora quantization candle from-scratch self-learning dora ai-agents mixture-of-experts quantization-aware-training sparse-attention llm large-language-model gguf multimodal-llm decoder-only grpo thinking-llm
-
Updated
Aug 21, 2026 - Rust