Files

• Changes
• MANIFEST
• META.json
• META.yml
• Makefile.PL
• README
• examples/01-model-info.pl
• examples/02-speculative-decoding.pl
• examples/download-model.pl
• lib/Lugh.pm
• lib/Lugh.xs
• lib/Lugh/Context.pm
• lib/Lugh/Graph.pm
• lib/Lugh/Inference.pm
• lib/Lugh/KVCache.pm
• lib/Lugh/LoRA.pm
• lib/Lugh/Model.pm
• lib/Lugh/Ops.pm
• lib/Lugh/Prompt.pm
• lib/Lugh/Prompt.xs
• lib/Lugh/Quant.pm
• lib/Lugh/Quant.xs
• lib/Lugh/RoPE.pm
• lib/Lugh/Speculative.pm
• lib/Lugh/Tensor.pm
• lib/Lugh/Tokenizer.pm
• lib/ppport.h
• lib/typemap
• t/0000-load.t
• t/0001-tensor.t
• t/0002-model.t
• t/0003-tokenizer.t
• t/0004-inference.t
• t/0005-tensor-ops.t
• t/0006-integration.t
• t/0007-generate.t
• t/0008-kvcache.t
• t/0009-performance.t
• t/0010-multi-arch.t
• t/0011-prompt.t
• t/0012-prompt-integration.t
• t/0013-backend.t
• t/0014-memory-pool.t
• t/0015-batch.t
• t/0016-edge-cases.t
• t/0017-sample-topk.t
• t/0018-inference-methods.t
• t/0019-model-tensors.t
• t/0020-lora-interface.t
• t/0021-lora-forward.t
• t/0022-lora-cache.t
• t/0023-lora-pool.t
• t/0024-lora-batch.t
• t/0025-quant.t
• t/0026-rope.t
• t/0027-rope-integration.t
• t/0028-unified-forward.t
• t/0029-speculative.t
• t/0030-srand.t
• t/0031-forward-all.t
• t/0032-speculative-reproducibility.t
• t/0033-error-handling.t
• t/0034-backend-validation.t
• t/0035-lora-advanced.t
• t/1000-memory-context.t
• t/1001-memory-model.t
• t/1002-memory-tokenizer.t
• t/1003-memory-inference.t
• t/1004-memory-kvcache.t
• t/1005-memory-pool.t
• t/1006-memory-speculative.t
• t/1007-memory-tensor.t
• t/1008-memory-lora.t
• t/1009-memory-rope.t
• t/data/MANIFEST.arch
• t/data/test-lora.gguf
• t/data/test-lora.safetensors
• t/data/test-model.gguf
• t/data/tiny-bloom.gguf
• t/data/tiny-falcon.gguf
• t/data/tiny-gemma2.gguf
• t/data/tiny-gpt2.gguf
• t/data/tiny-llama.gguf
• t/data/tiny-phi3.gguf
• t/data/tiny-qwen2.gguf
• t/data/tiny-starcoder.gguf
• t/manifest.t
• t/pod-coverage.t
• t/pod.t
• xt/boilerplate.t