lightonai/LightOnOCR-2-1B

#34 opened about 1 month ago by

jojo71

Excessive use of HTML in Output

❤️ 1

#33 opened about 2 months ago by

abimaelmartell

RLVR Strategy Suggestions

#30 opened 2 months ago by

TheOfficialAJ

Accuracy decreases a lot in gguf conversion

#28 opened 3 months ago by

fhaDL

How it is compared in bahchmarks with new model Paddle ocr 1.5 ?

#27 opened 3 months ago by

dibu28

How do you calculate inference speed on OlmOCR-Bench?

👍 1

#26 opened 3 months ago by

Piperino

Throughput is slower than PadlleOCR-VL

👍 1

#25 opened 3 months ago by

TechNetiums

Regarding reasoning capability on image

#24 opened 3 months ago by

Ajayan

The exact vocab size of the model

#22 opened 3 months ago by

abdullahamlwakeb

Future suggestion about using available text layers

#21 opened 3 months ago by

jondecker76

Prompt of Demo web

#20 opened 3 months ago by

vanhdz2611

whats the best settings to run using vllm at very high concurrency

➕ 3

#19 opened 3 months ago by

markwitt1

Fine-tuning for table bounding box extraction

#18 opened 3 months ago by

TheOfficialAJ

Fine-tune for structured extraction

👍 1

#17 opened 3 months ago by

Glider95

x,y coordinates

🚀 1

#16 opened 3 months ago by

Superdooperhero

Incredible Model, Thank You!

❤️ 4

#15 opened 3 months ago by

md-1415

Amazing model

👍❤️ 2

#14 opened 3 months ago by

leniad

Experience on finetuning for specific language

#13 opened 3 months ago by

Huy227

Fix typo in model architecture class name

#12 opened 3 months ago by

Xenova

Apple users fear not you can use MLX too!

🚀🔥 6

#11 opened 3 months ago by

pherber3

LightOnOCR-2-1B with Samaritan Hebrew

❤️ 3

#10 opened 3 months ago by

johnlockejrr

any way to get a shrunk output to reduce tokens?

3

#9 opened 3 months ago by

Mr-Vader

Muliltipage documents

#8 opened 3 months ago by

jondecker76

Tables as markdown instead of HTML

5

#6 opened 3 months ago by

jondecker76

Remove page numbers from the output