Downloads · 30 days
10
100% of all-time downloads
perception-slm-image-v0
LNTTushar/perception-slm-image-v0
perception-slm-image-v0 is a machine learning model from LNTTushar. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for perception-slm. The card lists the license as apache-2.0.
Small, from-scratch image-understanding model (Perception SLM, Phase 2 / image v0). Encoder → connector (resampler) → tiny decoder, trained with Stage-2 alignment and Stage-3 LoRA instruction tuning. Config: imagev0gpu.
All-time downloads
10
Public
Repo size
512 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt512 MB · 100%
At a glance
- Library
- perception-slm
- License
- apache-2.0
- Access
- Public
- Created
- Sep 17, 2026
- Updated
- Sep 17, 2026
- SHA
- 05a118cb
Model card
From the Hugging Face model README
LNTTushar/perception-slm-image-v0
Small, from-scratch image-understanding model (Perception SLM, Phase 2 / image v0).
Encoder → connector (resampler) → tiny decoder, trained with Stage-2 alignment and
Stage-3 LoRA instruction tuning. Config: image_v0_gpu.
Held-out metrics
| metric | value |
|---|---|
| loss | 2.0373 |
| token_acc | 0.555 |
| exact_match | 0.0 |
| n | 294 |
Files
model.pt— checkpoint (model_state+ training metadata)config.yaml— the exact config used to build the modelmodel_int8.pt— int8-quantized weights for CPU/offline (if uploaded)
Load
from huggingface_hub import hf_hub_download
import torch
ckpt = torch.load(hf_hub_download("LNTTushar/perception-slm-image-v0", "model.pt"), map_location="cpu")
# rebuild ImageVLM.from_config(config) then load_state_dict(ckpt["model_state"])
Built with the perception-slm repo; see its RESULTS.md.
- Library
- perception-slm
- License
- apache-2.0
- Created
- Sep 17, 2026
- Updated
- Sep 17, 2026