This is GGUF quantization of
BLR2/Qwen3.5-9B-Eagle3-ShareGPT ,trained on ShareGPT.
You can use it as Eagle3 speculative decoding drafter for any
Qwen3.5-9B quantization.
Now that
llama.cpp has merged support for
Eagle3 in Qwen, you can use any build past
b9723.
