Qwen3.8-27B-pi-adapted

Try Try

Qwen3.8-27B-pi-adapted has 159M parameters (159M active per token) and takes 638 MB on disk. See where its parameters live, layer by layer, and what hardware it needs.

At a glance

harness-hub/Qwen3.8-27B-pi-adapted
Repo createdNov 16, 2026updated Nov 18, 2026
Model typeDenseFull attention
InputsText
Total parameters159M
Active per token159M100% of the model
ExpertsNone (dense)
Max context (from config)—
Layers—the files show 64, which doesn't match; layer details hidden
On disk638 MB1 file
PrecisionFP32 (100%)
QuantizationNoneoriginal precision (FP32)
Fewest GPUs1× RTX 4060a single 8 GB card · weights only
FamilyAdapter of Qwen/Qwen3.8-27B
Licenseapache-2.0
GitHubNot linked