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: 2,912 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 | from __future__ import annotations
from collections.abc import Callable
from typing import TYPE_CHECKING
if TYPE_CHECKING:
from .api import PredictRequest
from .backend import EncodedRequest
import re
from . import grounding, image_focus
VIEW_ID = "pdf-source-only-scalars-v1"
_PAGE_SCOPE = re.compile(
"\\b(?:page\\s+(?:[1-4]|one|two|three|four)|(?:first|second|third|fourth|1st|2nd|3rd|4th)\\s+page)\\b",
re.IGNORECASE,
)
_EXCLUDED_SCOPE = re.compile(
"\\b(?:not|ignore|exclude|excluding|except|without|rather than)\\b[^.;:\\n]*$",
re.IGNORECASE,
)
def _page_scoped(instructions: object) -> bool:
return isinstance(instructions, str) and any(
(
not _EXCLUDED_SCOPE.search(instructions[: match.start()])
for match in _PAGE_SCOPE.finditer(instructions)
)
)
def prepare(
request: PredictRequest,
encoded: EncodedRequest,
encode: Callable[..., EncodedRequest],
*,
enabled: bool,
) -> EncodedRequest:
if not enabled or not request.documents:
return encoded
(record, forecasts, tasks, typed, eos) = encoded
active = {name for (name, _, _) in tasks}
selected = {
name: spec
for (name, spec) in request.fields.items()
if name in active
and spec.get("type") == "number"
and (spec.get("source_only") is True)
and (not spec.get("knowledge"))
and (not grounding.state_only(spec))
and _page_scoped(spec.get("instructions", ""))
}
originals = record.get("_original_images", ())
pdf_pages = originals[len(request.images) :]
if (
not selected
or not pdf_pages
or (
not any(
(image_focus.focus_document(page) is not page for page in pdf_pages)
)
)
):
return encoded
child_request = request.model_copy(update={"fields": selected})
(child, _, replacements, child_typed, child_eos) = encode(
child_request,
identifier_view=True,
image_views=lambda: originals,
image_view_key=VIEW_ID,
)
if child_typed is not None or child_eos != eos:
raise ValueError("A literal PDF number view must retain the scalar AR contract")
by_name = {name: (name, ids, media) for (name, ids, media) in replacements}
if set(by_name) != set(selected):
raise ValueError(
"PDF number detail tasks differ from the selected public fields"
)
updated = [by_name.get(name, (name, ids, media)) for (name, ids, media) in tasks]
salts = dict(record.get("_ar_task_salts", {}))
for name in selected:
record["_ar_constraints"][name] = child["_ar_constraints"][name]
salts[name] = child["_cache_salt"]
record.update(
_ar_tasks=updated, _ar_task_salts=salts, _ar_prefix=[], _ar_group_prefixes=[]
)
return (record, forecasts, updated, typed, eos)
|