100% Private
No Signup
Free Forever
One of 64 free AI tools by Mahmoud Zalt.
Free AI Image Captioner & Alt-Text Generator
A free, private AI image captioner that describes any image and writes accessibility-ready alt text without sending your files anywhere. It runs the vit-gpt2-image-captioning vision-language model directly in your browser using Hugging Face Transformers.js, the library for state-of-the-art machine learning on the web, so your images never leave your device. Transformers.js runs the same pretrained models as the Python transformers library, executing them client-side through ONNX Runtime on WebAssembly (WASM) with optional WebGPU acceleration. Drag and drop a JPG, PNG, WebP, GIF, or BMP, load the model once (about 250MB, cached for next time), and get a natural-language caption you can copy as plain text or as a complete <img alt="..."> HTML snippet. Ideal for developers improving accessibility, content teams writing alt text at scale, and anyone who wants quick image-to-text descriptions without a cloud API or account.
Free and provided as is, without warranty. Use at your own risk. Terms
Free image captioning online, privately in your browser
This free AI image captioner turns any photo, screenshot, or graphic into a clear one-line description and a ready-to-paste HTML alt attribute. Everything runs locally using Hugging Face Transformers.js, the library for state-of-the-art machine learning on the web, so your images never leave your device and you never need an account or API key.
Drag and drop an image, load the model once, and get a caption you can copy as plain text or as a complete <img alt="..."> snippet. It is built for developers, content teams, and anyone who wants fast, private image-to-text descriptions without sending files to a cloud service.
How it works: Transformers.js and vit-gpt2-image-captioning in the browser
Transformers.js runs the same pretrained models as the Python transformers library, but directly in your browser with no server needed. It executes models through ONNX Runtime on WebAssembly (WASM), with WebGPU acceleration when your browser supports it, so inference happens entirely on your own hardware.
The captioning model is vit-gpt2-image-captioning (Xenova/vit-gpt2-image-captioning), a vision-language architecture that combines a Vision Transformer (ViT) image encoder with a GPT-2 text decoder. The ViT reads the pixels into a visual representation and GPT-2 generates a natural-language sentence from it, which the tool then cleans up into a usable caption or alt attribute.
Why in-browser image captioning beats cloud APIs
Most image captioning services upload your files to a remote server, require sign-up, and meter usage by request. This tool flips that model: the AI runs on your machine via Transformers.js, so there are no per-image fees, no rate limits, and no privacy tradeoffs. You can verify nothing is uploaded by watching the DevTools Network tab while you caption.
After the one-time model download (about 250MB, cached by your browser), captioning works fast and even offline. That makes it practical for batches of images and for sensitive screenshots you would rather not hand to a third-party API. Both Transformers.js and the vit-gpt2-image-captioning model are open source under the Apache 2.0 license.
More Free Tools
More than 20 free AI tools.
Better accessibility and image SEO with descriptive alt text
Alt text is essential for accessibility: screen readers announce it to users who cannot see the image, and it appears when an image fails to load. Search engines also rely on alt text to understand and rank images, so meaningful descriptions help both people and discoverability in image search.
Use the generated caption as a strong first draft, then refine it for context. Pairing automated image-to-text captioning with a quick human review is the fastest way to ship accurate, accessible alt text across an entire site without paying for a cloud vision API.
Who actually needs alt text at scale
Accessibility audits are a common trigger: WCAG 2.1 requires meaningful alt text for informative images, and both the ADA in the US and the EN 301 549 standard in the EU reference WCAG conformance directly, so a legal or procurement deadline often turns "we should add alt text" into "we need alt text on 2,000 product photos by Friday." Running a product catalog through this tool one image at a time is far faster than a person writing every description from scratch.
Documentation teams caption dozens of UI screenshots in a help center so screen-reader users get a real description instead of a filename, and content teams draft a starting caption for editorial images before a human editor tightens the wording for tone and context. In every case, the model's job is to produce a solid first draft, not the final, ship-ready copy.
How this compares to Google Cloud Vision and Azure Computer Vision
Google Cloud Vision's label and caption detection bills per image after a small free monthly allowance, typically working out to roughly 1 to 1.50 dollars per thousand images, and Azure Computer Vision's image analysis pricing is in a similar range. Both are capable, well-documented APIs, but every image has to be uploaded to that provider's servers to get a caption back, and the cost adds up quickly for a large photo library.
This tool runs a single open vision-language model, vit-gpt2-image-captioning, entirely on your device: no per-image charge, no API key, and no upload step. The tradeoff is capability, the large cloud vision APIs support additional features like object detection, landmark recognition, and multiple caption candidates that a lightweight browser model does not attempt, so for a one-off batch of alt text this tool wins on cost and privacy, while a product that needs rich, ongoing image analysis at scale may still want a cloud API.
How It Works
Drag and drop an image, or click to choose a JPG, PNG, WebP, GIF, or BMP file from your device.
Load the captioning model once (about 250MB); it downloads from the Hugging Face CDN and is cached in your browser for future visits.
The model reads the image locally and returns a one-line caption you can copy as plain text or as a ready-to-paste HTML alt attribute.
Need expert help with AI?
Looking for a specialist to help integrate, optimize, or consult on AI systems? Book a one-on-one technical consultation with an experienced AI consultant to get tailored advice.
Key Features
Privacy & Trust
Use Cases
Limitations
- The first load downloads about 250MB of vit-gpt2-image-captioning model files; expect a wait on slow connections, though it is cached by your browser afterward.
- Captions are short, single-sentence image-to-text descriptions: the model will not read text in the image, count objects precisely, or identify specific people.
- vit-gpt2-image-captioning is a general-purpose model trained on common scenes, so highly technical, abstract, or unusual images may get vague or imperfect captions.
- Performance depends on your device CPU, memory, and WebGPU support; very large images and older machines run more slowly than recent hardware.
Frequently Asked Questions
Is this image captioner free?
Yes, it is completely free with no limits, no per-image fees, and no rate caps. There is nothing to buy and no usage metering because the AI runs on your own device rather than on a paid cloud service. It is built on Hugging Face Transformers.js and the vit-gpt2-image-captioning model, both open source under the Apache 2.0 license, so the tool itself costs nothing to run.
Are my images uploaded to a server?
No. The entire captioning process happens locally in your browser using Hugging Face Transformers.js running on ONNX Runtime and WebAssembly. Your image is read into memory on your device and analyzed there, never transmitted. The only thing downloaded from the internet is the AI model itself, which is identical for every user and contains none of your data. You can confirm this in the DevTools Network tab while captioning.
Do I need to install anything or create an account?
No. There is no software to install, no browser extension, no account, no email, and no API key. Everything runs inside the web page you are already on. Transformers.js loads the model and runs inference entirely client-side, so you just drop in an image and get a caption.
How accurate are the captions?
The tool uses vit-gpt2-image-captioning, a widely used image-to-text model that pairs a Vision Transformer (ViT) image encoder with a GPT-2 text decoder. It produces solid one-line descriptions of common scenes, objects, and activities and is well suited for general captions and alt-text drafts. For important or domain-specific content you should review and lightly edit the output, since a general-purpose model can occasionally be vague or miss details.
Why is the first load slow, and how big is the model?
AI image captioning needs a vision-language model to run, and that model is about 250MB. The first time you caption an image the weights download once from the Hugging Face Hub CDN, which is the slow step on a fresh visit. After that your browser caches the files, so later captions start almost instantly and even work offline. We do not auto-download on page load so you stay in control of your bandwidth.
Does it work offline or on mobile?
Once the model has been downloaded and cached, captioning runs with no network connection because all inference executes on your device through WebAssembly. It also works on modern mobile browsers, though phones and tablets are slower than desktops and the initial 250MB download is heavier on cellular data. The only step that needs the internet is the one-time model download.
What is alt text and why does it matter?
Alt text is the text alternative for an image, set via the HTML alt attribute. Screen readers announce it to people who are blind or have low vision, browsers display it when an image fails to load, and search engines read it to understand and rank images. Good alt text improves both accessibility and image SEO, and this tool gives you a ready-to-paste <img alt="..."> starting point you can refine for context.
Can it read or transcribe text inside an image?
No. This is an image captioning model, not an OCR (optical character recognition) tool. vit-gpt2-image-captioning describes the overall scene and content of an image rather than extracting words from it, so it will not transcribe documents, screenshots of text, or signs. For pulling text out of images you would need a dedicated OCR tool instead.
Q&A SESSION
Got a quick technical question?
Skip the back-and-forth. Get a direct answer from an experienced engineer.