Downloads · 30 days
100
51% of all-time downloads
jinzex/Qwen3.5-35B-A3B-DSpark
Qwen3.5-35B-A3B-DSpark is a machine learning model from jinzex. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This is a DSpark drafter for Qwen/Qwen3.5-35B-A3B, trained with Model Optimizer.
Downloads · 30 days
100
51% of all-time downloads
All-time downloads
197
Public
Parameters
505M
2 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors1 GB · 100%
From the Hugging Face model README
This is a DSpark drafter for Qwen/Qwen3.5-35B-A3B, trained with Model Optimizer.
v1.0: original drafter trained on Nemotron Post-Training Dataset v2v1.1: continued fine-tuning on Qwen3.5 SWE rollout traces with a 32K training sequence lengthModel Optimizer PR #2340 · E2E Training
| Category | T=1 | T=0 |
|---|---|---|
| Coding | 4.3720 | 4.8880 |
| Humanities | 3.0543 | 3.6332 |
| Math | 4.0852 | 4.5721 |
| Multilingual | 3.7786 | 4.1805 |
| QA | 3.2111 | 3.7132 |
| RAG | 4.0790 | 4.4432 |
| Reasoning | 3.6061 | 3.9967 |
| Roleplay | 2.5711 | 3.3469 |
| STEM | 3.3168 | 3.8496 |
| Summarization | 3.5660 | 4.0488 |
| Writing | 2.7125 | 2.9998 |
| Overall AL | 3.4866 | 3.9702 |