KBaba7/llama.cpp

1# llama.cpp/examples/speculative-simple2 3Demonstration of basic greedy speculative decoding4 5```bash6./bin/llama-speculative-simple \7    -m  ../models/qwen2.5-32b-coder-instruct/ggml-model-q8_0.gguf \8    -md ../models/qwen2.5-1.5b-coder-instruct/ggml-model-q4_0.gguf \9    -f test.txt -c 0 -ngl 99 --color \10    --sampling-seq k --top-k 1 -fa --temp 0.0 \11    -ngld 99 --draft-max 16 --draft-min 5 --draft-p-min 0.912```13