Image-Text-to-Text
MLX
Safetensors
qwen3_5
clairic
cloudflare
systemone
structured-output
information-extraction
multimodal
custom-code
conversational
4-bit precision
Instructions to use mlx-community/clairic-flash-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/clairic-flash-4bit with MLX:
# Make sure mlx-vlm is installed # pip install --upgrade mlx-vlm from mlx_vlm import load, generate from mlx_vlm.prompt_utils import apply_chat_template from mlx_vlm.utils import load_config # Load the model model, processor = load("mlx-community/clairic-flash-4bit") config = load_config("mlx-community/clairic-flash-4bit") # Prepare input image = ["http://images.cocodataset.org/val2017/000000039769.jpg"] prompt = "Describe this image." # Apply chat template formatted_prompt = apply_chat_template( processor, config, prompt, num_images=1 ) # Generate output output = generate(model, processor, formatted_prompt, image) print(output) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Pi
How to use mlx-community/clairic-flash-4bit with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "mlx-community/clairic-flash-4bit"
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "mlx-lm": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "mlx-community/clairic-flash-4bit" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent
How to use mlx-community/clairic-flash-4bit with Hermes Agent:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "mlx-community/clairic-flash-4bit"
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default mlx-community/clairic-flash-4bit
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use mlx-community/clairic-flash-4bit with OpenClaw:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "mlx-community/clairic-flash-4bit"
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:8080/v1 \ --custom-model-id "mlx-community/clairic-flash-4bit" \ --custom-provider-id mlx-lm \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
File size: 2,679 Bytes
ae97c5d | 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 | from __future__ import annotations
import math
import re
from typing import Any
import numpy as np
from PIL import Image, ImageChops
VIEW_ID = "party-name-header-lines-v1"
VIEW_INSTRUCTION = " The first image is the complete document. The second image enlarges its first printed lines. Use both views of the same source to read the complete requested name; preserve every digit in the name."
_ROLE = re.compile(
"\\b(?:seller|merchant|vendor|customer|bill[- ]?to|recipient|payee|payer|guest|organization|business|company|legal)\\b",
re.IGNORECASE,
)
_OTHER_VALUE = re.compile(
"\\b(?:address|number|identifier|code|date|row|label|amount)\\b", re.IGNORECASE
)
def fields(schema: dict[str, dict[str, Any]]) -> list[str]:
selected = []
for name, spec in schema.items():
instructions = spec.get("instructions", "")
if (
spec.get("type") != "text"
or spec.get("knowledge")
or (not isinstance(instructions, str))
):
continue
requested = re.split("[.\\n]", instructions, maxsplit=1)[0]
if (
re.search("\\bname\\b", requested, re.IGNORECASE)
and _ROLE.search(requested)
and (not _OTHER_VALUE.search(requested))
):
selected.append(name)
return selected
def header_detail(
image: Image.Image, *, max_pixels: int = 1600000
) -> Image.Image | None:
if image.mode != "RGB" or image.getpixel((0, 0)) != (255, 255, 255):
return None
ink = np.asarray(image.convert("L")) < 200
if np.mean(ink) > 0.22:
return None
edges = np.flatnonzero(np.diff(np.pad(ink.any(axis=1).astype(np.int8), (1, 1))))
bands = edges.reshape(-1, 2).tolist()[:6]
if not bands or any((not 8 <= bottom - top <= 32 for (top, bottom) in bands)):
return None
top = max(0, bands[0][0] - 24)
bottom = min(image.height, bands[-1][1] + 24)
block = image.crop((0, top, image.width, bottom))
bounds = ImageChops.difference(
block, Image.new("RGB", block.size, "white")
).getbbox()
if bounds is None:
return None
(left, right) = (max(0, bounds[0] - 24), min(block.width, bounds[2] + 24))
block = block.crop((left, 0, right, block.height))
scale = min(3.0, math.sqrt(max_pixels / (block.width * block.height)))
size = (max(1, int(block.width * scale)), max(1, int(block.height * scale)))
return block.resize(size, Image.Resampling.BICUBIC)
def views(image: Image.Image, *, max_pixels: int = 1600000) -> tuple[Image.Image, ...]:
detail = header_detail(image, max_pixels=max_pixels)
return (image, detail) if detail is not None else (image,)
|