Downloads · 30 days
7
22% of all-time downloads
Qwen3.5-9B-OPSD-PI-Strong-ckpt15
LSW142857/Qwen3.5-9B-OPSD-PI-Strong-ckpt15
Qwen3.5-9B-OPSD-PI-Strong-ckpt15 is a image-text-to-text model from LSW142857. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.
This is the self-contained Hugging Face model obtained by merging the final Strong-PI OPSD LoRA adapter (checkpoint 15) into Qwen/Qwen3.5-9B with alpha / rank = 128 / 64 = 2.0.
All-time downloads
32
Public
Parameters
9.7B
19.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors19.3 GB · 100%
Parameter types
How the weights are stored.
BF169.7B · 100%
Try a prompt
Base models
Model card
From the Hugging Face model README
Qwen3.5-9B OPSD-PI Strong checkpoint 15
This is the self-contained Hugging Face model obtained by merging the final
Strong-PI OPSD LoRA adapter (checkpoint 15) into Qwen/Qwen3.5-9B with
alpha / rank = 128 / 64 = 2.0.
The run used stage-adaptive privileged information over 512 coding-agent trajectories. PI was routed over EXPLORE, REPRODUCE, DIAGNOSE, EDIT, and VERIFY; SUBMIT remained executable but was masked from the distillation loss.
Download
hf download LSW142857/Qwen3.5-9B-OPSD-PI-Strong-ckpt15 \
--repo-type model \
--local-dir Qwen3.5-9B-OPSD-PI-Strong-ckpt15
The four safetensors shards contain the complete base model plus the merged adapter, so no separate base-model download is required. The model is intended for text coding-agent evaluation. The training-time standalone MTP parameters are not part of the LoRA-only adapter; use target-model decoding (speculative decoding disabled) for the reported evaluation contract.
See MERGED_MODEL_MANIFEST.json for checkpoint provenance and shard checksums.
- Task
- Image-Text-to-Text
- Library
- transformers
- Type
- qwen3_5
- License
- apache-2.0
- Created
- Aug 15, 2026
- Updated
- Aug 17, 2026