Hugging Face
Search models, datasets, users...
Ctrl K
Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
btbtyler09
/
shrew-ocr-preview-FP8
Like
0
Image-Text-to-Text
Safetensors
English
granite4_vision
fp8
compressed-tensors
document-intelligence
structured-extraction
vision-language
ocr
rag
conversational
custom_code
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
Copy to bucket
main
shrew-ocr-preview-FP8
4.86 GB
Ctrl+K
Ctrl+K
1 contributor
History:
4 commits
btbtyler09
v0.5 card update: wording fixes (hardware wording, FP8 behaviour sentence, GGUF alias)
a23cf87
verified
6 days ago
.gitattributes
Safe
1.52 kB
initial commit
6 days ago
README.md
Safe
12.2 kB
v0.5 card update: wording fixes (hardware wording, FP8 behaviour sentence, GGUF alias)
6 days ago
SHA256SUMS
Safe
1.34 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
added_tokens.json
Safe
24 Bytes
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
chat_template.jinja
Safe
12.1 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
config.json
Safe
3.24 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
configuration.py
Safe
1.85 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
downsampling.py
Safe
6.05 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
generation_config.json
Safe
187 Bytes
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
merges.txt
Safe
917 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
model.safetensors
4.85 GB
xet
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
modeling.py
Safe
24.1 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
preprocessor_config.json
Safe
677 Bytes
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
processing.py
Safe
2.35 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
processor_config.json
Safe
286 Bytes
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
special_tokens_map.json
Safe
631 Bytes
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
tokenizer.json
Safe
7.15 MB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
tokenizer_config.json
Safe
18 kB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago
vocab.json
Safe
2.01 MB
v0.5 — W8A8 FP8 (compressed-tensors, calibrated on 480 real pages), v0.5: v0.4.1 plus two supervised fine-tuning stages and a one-direction repetition-loop edit (1,024 MLP output columns, 0.066 % of weights); OmniDocBench v1.6 overall 77.12 to 83.45; requires shrew-server >= 0.5.0
6 days ago