Downloads · 30 days
13
7% of all-time downloads
shisa-ai/Qwen3.6-35B-A3B-PARO-full4096-e5-packed
Qwen3.6-35B-A3B-PARO-full4096-e5-packed is a text generation model from shisa-ai. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This is the packed ParoQuant export for Qwen/Qwen3.6-35B-A3B, using the full4096-e5 calibration run.
Downloads · 30 days
13
7% of all-time downloads
All-time downloads
187
Public
Parameters
6B
20.5 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors20.5 GB · 100%
How the weights are stored.
I324.2B · 71%
From the Hugging Face model README
This is the packed ParoQuant export for Qwen/Qwen3.6-35B-A3B, using the full4096-e5 calibration run.
The packed artifact was produced from the legacy/original export with:
python3 scripts/strip_paro_safetensors.py \
--input-dir /models/qwen36-quant/Qwen3.6-35B-A3B-PARO-full4096-e5 \
--output-dir /models/qwen36-quant/Qwen3.6-35B-A3B-PARO-full4096-e5-packed \
--mode packed
Packed changes:
.weight fallback tensor where the same module has .qweightmodel.safetensors: 20,474,495,512 bytesThe legacy/original-format export is available separately at:
Canonical tx4/quality3 evaluation compares each candidate directly against the original BF16 HF model on the same scored token positions.
| Model | Format | Size GiB ↓ | BPW ↓ | PPL ↓ | ΔNLL ↓ | KL nats ↓ | Top-1 % ↑ |
|---|---|---|---|---|---|---|---|
| Original BF16 HF | HF safetensors | 66.966 | 16.435 | 6.5590 | +0.000000 | 0.000000 | 100.000 |
| PARO full4096-e5 packed (this repo) | packed safetensors | 19.068 | 4.680 | 6.6216 | +0.009506 | 0.034684 | 92.000 |
| PARO full4096-e5 unpacked/original-format | legacy safetensors | 21.686 | 5.322 | 6.6216 | +0.009506 | 0.034684 | 92.000 |
| GGUF UD-Q4_K_S | GGUF | 19.458 | 4.776 | 6.5783 | +0.003842 | 0.012800 | 94.999 |
| GGUF UD-Q4_K_M | GGUF | 20.614 | 5.059 | 6.5643 | +0.001718 | 0.010849 | 95.354 |
Column calculations:
2^30; for HF/PARO this sums .safetensors weight files only, and for GGUF this is the .gguf file size.35,000,000,000; the fixed 35B denominator is used only to make formats directly comparable.exp(mean NLL) over the canonical scored target positions only. Evaluation source: tx4/quality3 validation set, ctx=2048, stride=1023, 127 windows, 129,921 scored tokens.KL(P_original_BF16_HF || P_candidate) over the full next-token distribution; lower is better.This artifact requires a packed-aware ParoQuant-compatible loader/runtime; legacy loaders that expect duplicate fp16 fallback .weight tensors will not load this format.
See strip_paro_safetensors_report.json for the exact stripping report.