PaddleOCR-VL ยท GGUF (llama.cpp)

GGUF quantizations of PaddlePaddle/PaddleOCR-VL for llama.cpp / Ollama / LM Studio. Produced by axteam.

Files

File Size Notes
PaddleOCR-VL-Q4_K_M.gguf 0.3 GB 4-bit k-quant (medium) โ€” recommended size/quality balance.
PaddleOCR-VL-Q8_0.gguf 0.5 GB 8-bit โ€” closest to the original.

Converted with llama.cpp convert_hf_to_gguf.py (recognized as paddleocr (ERNIE-4.5 text backbone)); the original chat_template is embedded. The vision encoder is not included in these GGUF files โ€” this is the text-backbone LLM only, so it serves text generation, not the full OCR/image pipeline. For end-to-end OCR use the original model or the FP8 build axteam/PaddleOCR-VL-FP8-dynamic with vLLM.

Quantization method

These GGUF files are post-training quantization (PTQ) made with llama.cpp (convert_hf_to_gguf.py โ†’ llama-quantize): the BF16 weights are quantized block-wise (k-quants: per-block scales with mixed bit allocation), with no retraining. The original chat_template is embedded in each file.

Requirements

llama.cpp build โ‰ฅ b8110 โ€” architecture paddleocr needs โ‰ฅ b8110, and the commands on this card use -fa on (โ‰ฅ b6325). Load-checked here with b12005 (commit e0f142d) and b10724 on CPU โ€” the loader accepts the model, tokenizer and the -fa on / q8_0 KV-cache flags; no generation was run in that check.
Memory the chosen GGUF file (sizes in the table) plus KV cache for the context (-c); --cache-type-k/v q8_0 -fa on halves the KV cache.
Input text only โ€” no mmproj in this repo, so llama.cpp cannot do OCR / image input with it.
Ollama / LM Studio not tested here.

Run with llama.cpp

# pull one quant straight from the Hub and serve an OpenAI-compatible API on :8080
llama-server -hf axteam/PaddleOCR-VL-GGUF:Q4_K_M -c 8192 --jinja

# or download a single file and chat in the terminal
hf download axteam/PaddleOCR-VL-GGUF PaddleOCR-VL-Q4_K_M.gguf --local-dir .
llama-cli -m PaddleOCR-VL-Q4_K_M.gguf -c 8192 --jinja

Query the server:

curl http://localhost:8080/v1/chat/completions -H "Content-Type: application/json" -d '{
  "messages": [{"role": "user", "content": "๋Œ€ํ•œ๋ฏผ๊ตญ์˜ ์ˆ˜๋„๋Š” ์–ด๋””์ธ๊ฐ€์š”?"}],
  "max_tokens": 512
}'

Reduce KV-cache memory (optional, run-time)

KV-cache type is chosen at run time in llama.cpp (it is not baked into the file):

llama-server -hf axteam/PaddleOCR-VL-GGUF:Q4_K_M -c 8192 --jinja --cache-type-k q8_0 --cache-type-v q8_0 -fa on

Ollama

ollama run hf.co/axteam/PaddleOCR-VL-GGUF:Q4_K_M

LM Studio

Search for axteam/PaddleOCR-VL-GGUF in the model browser and pick a quant (e.g. Q4_K_M).

License

Quantized derivative of PaddlePaddle/PaddleOCR-VL (apache-2.0). Original weights & tokenizer belong to PaddlePaddle.


ํ•œ๊ตญ์–ด ์š”์•ฝ

PaddlePaddle/PaddleOCR-VL์˜ GGUF ์–‘์žํ™” ๋นŒ๋“œ (llama.cppยทOllamaยทLM Studio์šฉ) ยท ์ œ์ž‘ axteam

๊ฐœ์š”

ํ•ญ๋ชฉ ๋‚ด์šฉ
์›๋ณธ ๋ชจ๋ธ PaddlePaddle/PaddleOCR-VL
ํ˜•์‹ GGUF (llama.cpp)
๋ณ€ํ™˜ ๋„๊ตฌ llama.cpp convert_hf_to_gguf.py โ†’ llama-quantize
ํฌํ•จ ํ…์ŠคํŠธ ๋ฐฑ๋ณธ(์–ธ์–ด ๋ชจ๋ธ)๋งŒ โ€” ๋น„์ „ ์ธ์ฝ”๋” ๋ฏธํฌํ•จ, ์ด GGUF๋กœ๋Š” OCR ๋ถˆ๊ฐ€
ํŒŒ์ผ ์ˆ˜ 2๊ฐœ

์–‘์žํ™” ์ƒ์„ธ

  • ์–‘์žํ™” ์ ˆ์ฐจ: PTQ(ํ•™์Šต ํ›„ ์–‘์žํ™”, ์žฌํ•™์Šต ์—†์Œ), llama.cpp ๋ธ”๋ก ๋‹จ์œ„ ์–‘์žํ™”(k-quant Q*_K, Q8_0)
  • chat template: GGUF ํŒŒ์ผ์— ๋‚ด์žฅ
  • ํ’ˆ์งˆ: ๋ฏธ์ธก์ • โ€” ์•„๋ž˜ ํŒŒ์ผ๋ณ„ ์„ค๋ช…์€ llama.cpp ์–‘์žํ™” ๋‹จ๊ณ„์˜ ์ผ๋ฐ˜์  ๊ฒฝํ–ฅ

์ œ๊ณต ํŒŒ์ผ

ํŒŒ์ผ ํฌ๊ธฐ ์„ค๋ช…
PaddleOCR-VL-Q4_K_M.gguf 0.3 GB 4๋น„ํŠธ k-quant โ€” ํฌ๊ธฐยทํ’ˆ์งˆ ๊ท ํ˜• (๊ถŒ์žฅ)
PaddleOCR-VL-Q8_0.gguf 0.5 GB 8๋น„ํŠธ โ€” ์›๋ณธ์— ๊ฐ€์žฅ ๊ฐ€๊นŒ์›€

์š”๊ตฌ์‚ฌํ•ญ

  • llama.cpp: b8110 ์ด์ƒ (paddleocr ๊ตฌ์กฐ ์ง€์› b8110๋ถ€ํ„ฐ, ์˜ˆ์‹œ ๋ช…๋ น์˜ -fa on ์˜ต์…˜ b6325๋ถ€ํ„ฐ)
  • ๋กœ๋”ฉ ํ™•์ธ: b12005(์ปค๋ฐ‹ e0f142d)์™€ b10724์—์„œ CPU๋กœ ๋ชจ๋ธยทํ† ํฌ๋‚˜์ด์ € ๋กœ๋”ฉ๊ณผ -fa onยทq8_0 KV ์บ์‹œ ์˜ต์…˜ ์ธ์‹๋งŒ ํ™•์ธ (์ƒ์„ฑ์€ ์‹คํ–‰ ๋ฏธํ™•์ธ)
  • ๋ฉ”๋ชจ๋ฆฌ: ์„ ํƒํ•œ GGUF ํŒŒ์ผ ํฌ๊ธฐ + ์ปจํ…์ŠคํŠธ(-c)๋งŒํผ์˜ KV ์บ์‹œ (KV ์บ์‹œ ์ค„์ด๋Š” ๋ฐฉ๋ฒ•์€ ์•„๋ž˜ ์‚ฌ์šฉ๋ฒ• ์ฐธ๊ณ )
  • ์ž…๋ ฅ: ํ…์ŠคํŠธ ์ „์šฉ(mmproj ์—†์Œ) โ€” OCR์€ ์›๋ณธ ๋ชจ๋ธ์ด๋‚˜ FP8 ๋นŒ๋“œ(axteam/PaddleOCR-VL-FP8-dynamic)๋ฅผ vLLM์œผ๋กœ ์‹คํ–‰
  • OllamaยทLM Studio: ์ œ์ž‘ ์ธก์—์„œ ์‹คํ–‰ ๋ฏธํ™•์ธ

์‚ฌ์šฉ๋ฒ•

  • ํŒŒ์ผ ๋ฐ›๊ธฐ: hf download axteam/PaddleOCR-VL-GGUF PaddleOCR-VL-Q4_K_M.gguf --local-dir . (์•„๋ž˜ -m ๋ช…๋ น์— ํ•„์š”, ๋‹ค๋ฅธ ์–‘์žํ™”๋Š” ํŒŒ์ผ๋ช…์˜ Q4_K_M๋งŒ ๋ณ€๊ฒฝ)
  • ์„œ๋ฒ„ ์‹คํ–‰: llama-server -hf axteam/PaddleOCR-VL-GGUF:Q4_K_M -c 8192 --jinja โ€” OpenAI ํ˜ธํ™˜ API
  • ํ„ฐ๋ฏธ๋„ ๋Œ€ํ™”: llama-cli -m PaddleOCR-VL-Q4_K_M.gguf -c 8192 --jinja
  • Ollama: ollama run hf.co/axteam/PaddleOCR-VL-GGUF:Q4_K_M
  • LM Studio: ๋ชจ๋ธ ๊ฒ€์ƒ‰์—์„œ axteam/PaddleOCR-VL-GGUF ์ž…๋ ฅ ํ›„ ์–‘์žํ™” ๋‹จ๊ณ„ ์„ ํƒ
  • KV ์บ์‹œ ์ ˆ๊ฐ(์„ ํƒ): --cache-type-k q8_0 --cache-type-v q8_0 -fa on โ€” KV ์บ์‹œ ์•ฝ ์ ˆ๋ฐ˜, GGUF ํŒŒ์ผ๊ณผ ์ƒ๊ด€์—†์ด ์‹คํ–‰ ์‹œ ์ง€์ •

๋ผ์ด์„ ์Šค

  • ๋ผ์ด์„ ์Šค: Apache-2.0 โ€” ์›๋ณธ ๋ชจ๋ธ์˜ ๋ผ์ด์„ ์Šค๋ฅผ ๊ทธ๋Œ€๋กœ ๋”ฐ๋ฆ„
  • ๊ถŒ๋ฆฌ: ์›๋ณธ ๊ฐ€์ค‘์น˜ยทํ† ํฌ๋‚˜์ด์ €๋Š” PaddlePaddle(Baidu) ์†Œ์œ , ์ด ๋ฆฌํฌ๋Š” ์–‘์žํ™” ํŒŒ์ƒ๋ฌผ
Downloads last month
226
GGUF
Model size
0.5B params
Architecture
paddleocr
Hardware compatibility
Log In to add your hardware

4-bit

8-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for axteam/PaddleOCR-VL-GGUF

Quantized
(9)
this model

Collection including axteam/PaddleOCR-VL-GGUF