AI models & inference

Generate one image from a text prompt: app icon, brand logo, character mascot or illustration. POST JSON {description, type?, style?, aspect_ratio?}. type is icon, logo, mascot or illustration (default icon; only icon gets app-icon framing and mask safe margins). style is flat, metallic, 2d-game, clay, 3d, gradient or glassmorphism. aspect_ratio is 1:1, 4:3 or 16:9. Returns image_key and preview_url. No account required.

Complete parcel intelligence: owner, situs, assessed value, classification, property tax and tax-sale (delinquency) status, full sale history and deeds, census tract, CDBG low/mod status, and every qualifying funding program with its cited rule.

The full record for one of the 1,326 specific Texas Hold'em starting hands: equity against modelled ranges, position-by-position opening guidance, exact head-to-head figures, and the exact card-removal effect of holding these specific two cards.

Live streaming speech-to-text for voice agents: buy a prepaid session, stream mic or call audio over WebSocket, get partial transcripts as the speaker talks. No account, no API key, no per-minute minimum — flat USDC on Base. Multilingual with mid-sentence code-switching. This SKU: 5 minutes connected time at $0.06 flat.

The full record for one of the 1,326 specific Texas Hold'em starting hands: equity against modelled ranges, position-by-position opening guidance, exact head-to-head figures, and the exact card-removal effect of holding these specific two cards.

AI video generation via fal.ai — default model: ByteDance Seedance 2.0. Alternatives: fal-ai/sora-2/text-to-video, fal-ai/veo3.1. Dynamic pricing per request: at-cost fal.ai rate + 15% margin (floor $0.05); the advertised price is the Seedance 5s reference quote — the actual quote is returned per request. Supports text-to-video and image-to-video.

Checks one phone number offline: normalises it to E.164, works out its country (including the twelve calling codes shared by several countries, such as +1 US/Canada/Caribbean) and validates the national number length against that country's numbering plan. Returns the tel: URI and the national dialling form. Numbering data from Google's libphonenumber. No network call and no API key, so it cannot fail because a third party is down.

MCP server exposing 69 tools - LLM completion, read-only EVM tooling and pure crypto utilities. initialize and tools/list are free so any client or directory can enumerate the tools; only tools/call is charged, a flat $0.001 per call.

Roll dice with any number of faces and get every roll plus the total. Uses the platform CSPRNG with rejection sampling, so every face is equally likely - a plain modulo is biased. Standard dice notation: 3d6, 1d20.

One two-sentence horror story, in English or Spanish, by theme: domestic, technological, cosmic or folk. Returned SPLIT into setup and payoff as well as joined, so a narrating agent can deliver the first line, pause, and then land the second. Takes parameters, returns an id to avoid repeats, and accepts a seed for a reproducible answer.

Pick one or more items at random from a list you send, with or without repeats. The list is yours, so this is the unbiased chooser an agent needs when it has options and no reason to prefer any.

Transcribes any video URL to text with Whisper v3 large. Auto-detects 90+ languages and offers a translate-to-English mode. 60-min / 500MB max. Same backend as video-to-text under a clearer slug. Use it for video transcription, video to audio transcription, or video-to-text. For speaker labels, use the speaker-diarize endpoint.

Check robots.txt, sitemap, Link headers, markdown access, llms.txt, and AI content-signal hints for agent-facing discoverability.

Product hero loop prompt: the compiled prompt and render plan for a 5s orbit loop on seedance-1.5-pro

Faceless explainer prompt: the compiled prompt and render plan for a 10s 9:16 3-shot explainer on seedance-1.5-pro

Answer questions about an image or describe what it shows (Llama 3.2 11B vision), fixed price $0.008 USDC. Use this when the input is an image; use /chat for text-only questions. POST JSON body {image (required, base64 or data:image/... URI, max 4 MB decoded, PNG/JPEG/GIF/WEBP), prompt (optional question, defaults to describing the image), max_tokens (optional 1-512, default 256)}. Returns {text}. Unpaid calls get 402; paid calls settle only on success.

The deep council: 4 different AI models, two rounds. Each seat answers, then sees the other seats' positions and revises - so you get a verdict that survived cross-examination, with changed_after_debate on each seat and the dissent that held. For decisions where being wrong is expensive.

LLM chat completion with no API key or signup (fast open model, meta-llama/llama-3.3-70b-instruct). OpenAI-compatible: set your SDK base URL to https://keyless.run/fast/v1. Flat $0.002 per call, up to 1500 output tokens. Charged only on success.

Use when the user is choosing between 2 or 3 Dutch used cars: plates side by side with make, model, age, APK expiry, insurance, open recall, odometer verdict, import and taxi history, catalogue price, CO2 and power, the best per metric and red flags. Input: Dutch licence plates. Source: RDW.

OpenAI-compatible LLM inference and chat completions across seven open-weight reasoning, coding, and vision models. Choose Kimi K3, Qwen3.6, Qwen3.8, GLM-5.2, GLM-5.3, GLM-5.3-Flash, or DeepSeek V4.1 Flash. Use them to generate text or JSON, analyze images, and call client-provided tools. Supports up to 4 images and selectable reasoning controls. Pricing is quoted dynamically from the prompt and max_tokens.

Initiate an AI-powered outbound phone call via Bland.ai.

Semantic search building block: turns text into a 1024-number vector for RAG retrieval, dedup, and clustering (bge-m3), fixed price $0.005 USDC. Use this when comparing meaning between texts, not for answers (use /chat) or translation (use /translate). POST JSON body {text (required, max 8000 chars)}. Returns {embedding, dimensions}. Unpaid calls get 402; paid calls settle only on success.

Language conversion across 100+ pairs with auto-detected source unless given (m2m100), fixed price $0.005 USDC. Use this for changing languages only; use /chat for rewriting within one language. POST JSON body {text (required, max 2000 chars), target_lang (required code), source_lang (optional code, default en)}. Returns translated {text}. Unpaid calls get 402; paid calls settle only on success.

Quick opinion triage: labels text POSITIVE or NEGATIVE with confidence plus all scores (DistilBERT SST-2), fixed price $0.005 USDC. Use this for sentiment only, not nuanced moderation or answers (use /chat). POST JSON body {text (required, max 2000 chars)}. Returns {label, score, scores}. Unpaid calls get 402; paid calls settle only on success.

General question answering, summarization, and short-form generation for TEXT input (Llama 3.1 8B instruct), fixed price $0.01 USDC. Use this when you already have text; use /transcribe for audio and /caption for images. POST JSON body {prompt (required, max 2000 chars), max_tokens (optional 1-512, default 256)}. Returns {text}. Unpaid calls get 402; paid calls settle only on success.

Automatically select a suitable open-weight model from Kimi K3, Qwen3.6-35B-A3B, and GLM-5.2, then generate a chat completion. The exact price is calculated from the selected model, prompt, and requested maximum completion tokens.

Prompt injection detection and content moderation for AI agents: flags hidden instructions aimed at an LLM (none / suspicious / malicious + score) and harmful content (hate, violence, sexual, harassment, self-harm, illegal). Any language, JSON output.

tts-1: OpenAI TTS 1 text to speech, paid per call in USDC. OpenAI list $15 per 1M characters, plus $0.0005 (Base) or $0.0005 (Solana) per call; up to 4096 characters; voices: alloy, ash, coral, echo, fable, onyx, nova, sage, shimmer. Returns the audio bytes (mp3 by default). Rates: https://openai.mm.family/x402/pricing

tts-1-hd: OpenAI TTS 1 HD text to speech, paid per call in USDC. OpenAI list $30 per 1M characters, plus $0.0005 (Base) or $0.0005 (Solana) per call; up to 4096 characters; voices: alloy, ash, coral, echo, fable, onyx, nova, sage, shimmer. Returns the audio bytes (mp3 by default). Rates: https://openai.mm.family/x402/pricing

text-embedding-ada-002: OpenAI Text Embedding Ada 002 embeddings, paid per call in USDC. OpenAI list $0.1 per 1M input tokens, plus $0.0005 (Base) or $0.0005 (Solana) per call; minimum $0.001 per call; up to 8192 tokens per input. Standard OpenAI body. Rates: https://openai.mm.family/x402/pricing