Harald Hoyer
5ee2f65337
Serve only Qwen3.6-27B; remove the unused 35B-A3B preset. Tuning: - Move model-specific keys (spec-type, sampling temp/top-p/top-k/min-p) out of the [*] defaults into [Qwen3.6-27B] so they no longer leak onto other models; draft-mtp in particular only works on MTP-weighted models. - Drop the duplicate parallel key from [*]. - Bump ubatch-size 256 -> 512 for faster iGPU prefill on Strix Halo. - Add threads-batch = 16 to use all cores for prefill while keeping generation at threads = 8 under full GPU offload.
29 lines
718 B
INI
29 lines
718 B
INI
version = 1
|
|
|
|
[*]
|
|
flash-attn = on
|
|
parallel = 1
|
|
jinja = true
|
|
n-gpu-layers = 99
|
|
threads = 8
|
|
ubatch-size = 512
|
|
cache-type-k = bf16
|
|
cache-type-v = bf16
|
|
mmap = false
|
|
no-context-shift = true
|
|
chat-template-kwargs = {"preserve_thinking": true}
|
|
fit = on
|
|
c = 131072
|
|
|
|
[Qwen3.6-27B]
|
|
hf = unsloth/Qwen3.6-27B-MTP-GGUF:Q6_K
|
|
spec-type = draft-mtp
|
|
spec-draft-n-max = 6
|
|
threads-batch = 16
|
|
temp = 0.6
|
|
top-p = 0.95
|
|
top-k = 20
|
|
min-p = 0.0
|
|
parallel = 2
|
|
c = 524288
|
|
load-on-startup = true
|