Try 100+ AI models in one place. Ask up to four LLMs the same question side by side, then make images, videos, voice-overs, subtitles and translations from the same balance. Every answer shows what it cost.
Text and chat
Compare AI models
Ask one question and get answers from up to four models side by side, with what each cost and how fast it came.
AI translator
Translate between 34 common languages, or from English into 22 Indian languages.
Extract data from text
Pull names, dates and amounts out of an email, an invoice or a note, with how sure the model is.
AI text classifier
Ask your own questions about a message or review, such as “is it a complaint?”, and get each answer with its probability.
Images
AI image generator
Make images from text with FLUX, GPT-Image, Seedream and more; compare up to three models.
AI photo editor
Change a photo by describing it: “make it night” or “remove the people in the background”.
Remove background
Cut the subject out onto a transparent PNG for listings, profiles and slides.
Image upscaler
Make a small or soft photo bigger and sharper.
SVG generator
Describe an icon or illustration and get an SVG to edit in Figma, Illustrator or Inkscape.
Video
AI video generator
Text to video with Seedance, Runway and more: a short clip in one to three minutes.
Image to video
Bring a photo to life: describe what happens next and get a clip that starts from it.
Video upscaler
Get a short clip back at twice the size with sharper detail.
Video background remover
Cut a person or object out of a clip, ready to place over another background.
Audio and speech
Audio to text
Transcribe recordings, meetings or videos, with timings, as text or subtitles.
Subtitle generator
Make subtitles for a video, fix any line, and download SRT or VTT.
Text to speech
Pick a voice, listen, compare models and download the audio.
Run your own model instead
Deploy an open LLM as an API
Host Llama, Qwen, gpt-oss or Whisper on the cheapest GPU that fits, behind your own https API.
Check what a model needs
GPU memory and the cheapest machine for popular open models, live.
Benchmark it on real GPUs
Tokens per second and users per GPU before you deploy.
Frequently asked questions
What is an LLM playground?
A page where you try large language models without writing code: you type a prompt, pick models and compare their answers. GridPup's AI Playground lets you ask up to four models the same question side by side and shows what each answer cost and how fast it came.
Which AI models can I use?
Over 100 models: chat models from OpenAI (GPT), Anthropic (Claude), Google (Gemini) and xAI (Grok), image models such as FLUX, GPT-Image and Seedream, video models such as Seedance and Runway, and speech, translation and text-to-speech models.
Do I need a subscription to each AI company?
No. You use every model from one prepaid GridPup balance and pay only for what you run. The price of each answer, image or video is shown before or as you make it.
Do I need an account?
You can open the Playground and every tool without one. You sign in when you send your first request, and add funds when your balance runs out.