Instructions to use sandeeprdy1729/thrisul-parser with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sandeeprdy1729/thrisul-parser with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="sandeeprdy1729/thrisul-parser", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("sandeeprdy1729/thrisul-parser", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use sandeeprdy1729/thrisul-parser with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sandeeprdy1729/thrisul-parser"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sandeeprdy1729/thrisul-parser",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/sandeeprdy1729/thrisul-parser

SGLang

How to use sandeeprdy1729/thrisul-parser with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "sandeeprdy1729/thrisul-parser" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sandeeprdy1729/thrisul-parser",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "sandeeprdy1729/thrisul-parser" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sandeeprdy1729/thrisul-parser",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use sandeeprdy1729/thrisul-parser with Docker Model Runner:
```
docker model run hf.co/sandeeprdy1729/thrisul-parser
```

thrisul-parser / nemotron_parse /engine /convert /rich_table_fallback.py

sandeeprdy1729

Thrisul-Parser v0.1

37fb2b9 verified 16 days ago

Raw History Blame Contribute Delete

3.89 kB

# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: OpenMDW-1.1

"""Decode known rich inline commands for the legacy tabular fallback.
Used only after strict rich-grid parsing fails. This preserves unknown commands
and malformed arguments; it never invents cell contents or repairs table spans.
"""
import html
import re

try:
    from .hybrid_cleanup import decode_styles
except ImportError:
    from hybrid_cleanup import decode_styles

try:
    from .rich_table_reader import TAGS, LITERALS, argument
except ImportError:
    from rich_table_reader import TAGS, LITERALS, argument


def decode_known_inline(text, _depth=0):
    """Decode balanced, recognized inline arguments without exposing delimiters."""
    if _depth > 64:
        return text
    out = []
    i = 0
    known = set(TAGS) | set(LITERALS) | {"thead", "linebreak", "mbox", "hl", "colorbox"}
    while i < len(text):
        # Preserve escaped literal characters. In particular, an escaped
        # backslash followed by 'thead' must not become a formatting command.
        if text[i] == "\\" and i + 1 < len(text) and not text[i + 1].isalpha():
            literal = text[i + 1]
            out.append(
                "\\&#" + str(ord(literal)) + ";"
                if literal in "&%$#_*{}"
                else text[i : i + 2]
            )
            i += 2
            continue
        match = re.match(r"\\([A-Za-z]+)\{", text[i:])
        if not match or match[1] not in known:
            out.append(text[i])
            i += 1
            continue
        command = match[1]
        try:
            body, end = argument(text, i + len(command) + 1)
        except ValueError:
            out.append(text[i])
            i += 1
            continue
        if command in {"hl", "colorbox"}:
            try:
                from .semantic_styles import mark_open
            except ImportError:
                from semantic_styles import mark_open
            from bs4 import BeautifulSoup

            node = BeautifulSoup("<mark></mark>", "html.parser").mark
            if command == "colorbox":
                node["background"] = body
                try:
                    body, end = argument(text, end)
                except ValueError:
                    out.append(text[i:end])
                    i = end
                    continue
            out.append(
                mark_open(node) + decode_known_inline(body, _depth + 1) + "</mark>"
            )
        elif command in TAGS or command == "thead":
            inner = decode_known_inline(body, _depth + 1)
            tag = "nvp-header-cell" if command == "thead" else TAGS[command]
            out.append("<" + tag + ">" + inner + "</" + tag + ">")
        elif command in LITERALS and not body:
            # Entities cannot be mistaken for LaTeX row/column delimiters.
            out.append("\\&#" + str(ord(LITERALS[command])) + ";")
        elif command == "linebreak" and not body:
            out.append("<nvp-line-break></nvp-line-break>")
        elif command == "mbox" and not body:
            pass
        else:
            out.append(text[i:end])
        i = end
    return decode_styles("".join(out))


def restore_header_cells(html_text):
    from bs4 import BeautifulSoup

    # Restore explicit breaks after legacy whitespace normalization, which
    # otherwise collapses consecutive linebreak commands.
    html_text = html_text.replace("<nvp-line-break></nvp-line-break>", "<br>")
    if "<nvp-header-cell>" not in html_text:
        return html_text
    soup = BeautifulSoup(html_text, "html.parser")
    for cell in soup.find_all(["td", "th"]):
        markers = cell.find_all("nvp-header-cell")
        if markers:
            cell.name = "th"
            for marker in markers:
                marker.unwrap()
    return str(soup)

	# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
	# SPDX-License-Identifier: OpenMDW-1.1

	"""Decode known rich inline commands for the legacy tabular fallback.

	Used only after strict rich-grid parsing fails. This preserves unknown commands
	and malformed arguments; it never invents cell contents or repairs table spans.
	"""
	import html
	import re

	try:
	from .hybrid_cleanup import decode_styles
	except ImportError:
	from hybrid_cleanup import decode_styles

	try:
	from .rich_table_reader import TAGS, LITERALS, argument
	except ImportError:
	from rich_table_reader import TAGS, LITERALS, argument


	def decode_known_inline(text, _depth=0):
	"""Decode balanced, recognized inline arguments without exposing delimiters."""
	if _depth > 64:
	return text
	out = []
	i = 0
	known = set(TAGS) \| set(LITERALS) \| {"thead", "linebreak", "mbox", "hl", "colorbox"}
	while i < len(text):
	# Preserve escaped literal characters. In particular, an escaped
	# backslash followed by 'thead' must not become a formatting command.
	if text[i] == "\\" and i + 1 < len(text) and not text[i + 1].isalpha():
	literal = text[i + 1]
	out.append(
	"\\&#" + str(ord(literal)) + ";"
	if literal in "&%$#_*{}"
	else text[i : i + 2]
	)
	i += 2
	continue
	match = re.match(r"\\([A-Za-z]+)\{", text[i:])
	if not match or match[1] not in known:
	out.append(text[i])
	i += 1
	continue
	command = match[1]
	try:
	body, end = argument(text, i + len(command) + 1)
	except ValueError:
	out.append(text[i])
	i += 1
	continue
	if command in {"hl", "colorbox"}:
	try:
	from .semantic_styles import mark_open
	except ImportError:
	from semantic_styles import mark_open
	from bs4 import BeautifulSoup

	node = BeautifulSoup("<mark></mark>", "html.parser").mark
	if command == "colorbox":
	node["background"] = body
	try:
	body, end = argument(text, end)
	except ValueError:
	out.append(text[i:end])
	i = end
	continue
	out.append(
	mark_open(node) + decode_known_inline(body, _depth + 1) + "</mark>"
	)
	elif command in TAGS or command == "thead":
	inner = decode_known_inline(body, _depth + 1)
	tag = "nvp-header-cell" if command == "thead" else TAGS[command]
	out.append("<" + tag + ">" + inner + "</" + tag + ">")
	elif command in LITERALS and not body:
	# Entities cannot be mistaken for LaTeX row/column delimiters.
	out.append("\\&#" + str(ord(LITERALS[command])) + ";")
	elif command == "linebreak" and not body:
	out.append("<nvp-line-break></nvp-line-break>")
	elif command == "mbox" and not body:
	pass
	else:
	out.append(text[i:end])
	i = end
	return decode_styles("".join(out))


	def restore_header_cells(html_text):
	from bs4 import BeautifulSoup

	# Restore explicit breaks after legacy whitespace normalization, which
	# otherwise collapses consecutive linebreak commands.
	html_text = html_text.replace("<nvp-line-break></nvp-line-break>", "<br>")
	if "<nvp-header-cell>" not in html_text:
	return html_text
	soup = BeautifulSoup(html_text, "html.parser")
	for cell in soup.find_all(["td", "th"]):
	markers = cell.find_all("nvp-header-cell")
	if markers:
	cell.name = "th"
	for marker in markers:
	marker.unwrap()
	return str(soup)