Image-Text-to-Text
Transformers
Safetensors
qwen3_5
clairic
cloudflare
systemone
multimodal
structured-output
information-extraction
custom-code
conversational
Instructions to use Cloudflare/clairic-flash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Cloudflare/clairic-flash with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Cloudflare/clairic-flash") 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 AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Cloudflare/clairic-flash") model = AutoModelForMultimodalLM.from_pretrained("Cloudflare/clairic-flash", device_map="auto") 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?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Cloudflare/clairic-flash with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Cloudflare/clairic-flash" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Cloudflare/clairic-flash", "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/Cloudflare/clairic-flash
- SGLang
How to use Cloudflare/clairic-flash 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 "Cloudflare/clairic-flash" \ --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": "Cloudflare/clairic-flash", "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 "Cloudflare/clairic-flash" \ --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": "Cloudflare/clairic-flash", "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 Cloudflare/clairic-flash with Docker Model Runner:
docker model run hf.co/Cloudflare/clairic-flash
File size: 3,493 Bytes
dc7d8aa | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 | """Validate embedded PDFs and render their pages inside the request encoder."""
from __future__ import annotations
from typing import TYPE_CHECKING
if TYPE_CHECKING:
from PIL import Image
import base64
import binascii
import math
import re
from .embedded import MAX_IMAGE_BYTES, MAX_IMAGES
from pydantic import BaseModel, ConfigDict, StrictStr, model_validator
class ClairDocument(BaseModel):
model_config = ConfigDict(extra="forbid", frozen=True)
content_type: StrictStr
base64: StrictStr
@model_validator(mode="before")
@classmethod
def parse_document(cls, value: object) -> dict[str, str]:
if isinstance(value, str):
match = re.fullmatch(
r"data:application/pdf;base64,([A-Za-z0-9+/]*={0,2})",
value,
re.IGNORECASE,
)
if match is None:
raise ValueError("document must be an embedded PDF data URI")
value = {"content_type": "application/pdf", "base64": match[1]}
if type(value) is not dict or set(value) != {"content_type", "base64"}:
raise ValueError("document object must contain content_type and base64")
if value["content_type"] != "application/pdf" or not isinstance(
value["base64"], str
):
raise ValueError("document content type must be application/pdf")
try:
data = base64.b64decode(value["base64"], validate=True)
except (binascii.Error, ValueError) as error:
raise ValueError("document base64 is malformed") from error
if not data.startswith(b"%PDF-") or len(data) > MAX_IMAGE_BYTES:
raise ValueError("document must be a PDF within the decoded byte limit")
return {
"content_type": "application/pdf",
"base64": base64.b64encode(data).decode("ascii"),
}
def render(self, available_pages: int = MAX_IMAGES) -> list[Image.Image]:
import pypdfium2 as pdfium
from .orientation import orient_document
try:
pdf = pdfium.PdfDocument(base64.b64decode(self.base64))
try:
if not 0 < len(pdf) <= available_pages:
raise ValueError(
f"PDF pages and images together must not exceed {MAX_IMAGES}"
)
images = []
for index in range(len(pdf)):
page = pdf[index]
try:
width, height = page.get_size()
if not (
math.isfinite(width * height) and width > 0 and height > 0
):
raise ValueError("PDF page has invalid dimensions")
scale = min(2.0, math.sqrt(4_000_000 / (width * height)))
bitmap = page.render(scale=scale)
try:
images.append(
orient_document(bitmap.to_pil().convert("RGB"))
)
finally:
bitmap.close()
finally:
page.close()
return images
finally:
pdf.close()
except pdfium.PdfiumError as error:
raise ValueError(
"PDF cannot be opened or rendered; encrypted PDFs are unsupported"
) from error
|