LM/Qwen 3.8 27B

llama.cpp

Running

./llama.cpp/sycl/bin/llama-server \
  -hf unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_XL \
  --spec-type draft-mtp --spec-draft-n-max 2 -ngld 99 \
  -c 64000 --cache-type-k q8_0 --cache-type-v q8_0 \
  -fa on -ngl 99 --load-mode mlock -np 1 --jinja --reasoning-preserve -t 8

Benchmarking

./llama.cpp/sycl/bin/llama-bench \
  -hf unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_XL \
  -ctk q8_0 -ctv q8_0 -fa on -ngl 99 --load-mode mlock \
  -p 512,1024 -n 512

./llama.cpp/sycl/bin/llama-bench \
  -hf unsloth/Qwen3.8-27B-GGUF:UD-Q5_K_XL \
  -ctk q8_0 -ctv q8_0 -fa on -ngl 99 --load-mode mlock \
  -p 512,1024 -n 512

./llama.cpp/sycl/bin/llama-bench \
  -hf unsloth/Qwen3.8-27B-GGUF:UD-Q6_K_XL \
  -ctk q8_0 -ctv q8_0 -fa on -ngl 99 --load-mode mlock \
  -p 512,1024 -n 512

Experiment

hf models ls --search "Qwen3.8-27B" --apps llama.cpp --expand "downloads,likes,lastModified" --sort downloads --no-truncate --limit 10

hf models ls --tree -R -h unsloth/Qwen3.8-27B-GGUF

hf download unsloth/Qwen3.8-27B-GGUF --include "*UD-Q4*"
hf download unsloth/Qwen3.8-27B-GGUF --include "*UD-Q5*"
hf download unsloth/Qwen3.8-27B-GGUF --include "*UD-Q6*"
hf download unsloth/Qwen3.8-27B-GGUF --include "mmproj-BF16*"

tree -h ~/.cache/huggingface/hub/models--unsloth--Qwen3.8-27B-GGUF