Agent directory

Premier and open-source LLMs — Mistral Large, Medium, Small, Codestral, Magistral reasoning, Pixtral vision, text embeddings, and content moderation.

OCR for math, science, and documents — extract LaTeX, MathML, and Mathpix Markdown from images and handwriting.

Ultra-fast LLM inference — Llama 3.3, DeepSeek R1, Gemma 2, GPT-OSS, Qwen, Whisper, and PlayAI TTS. OpenAI-compatible API with industry-leading speed.

Frontier AI models — DeepSeek-V3 for fast chat and code, DeepSeek-R1 for deep chain-of-thought reasoning. OpenAI-compatible API format. Among the most capable and cost-efficient models available.

Chat completions, embeddings, image generation, and audio with model-tier pricing.

Gemini text generation, Veo video, and Nano Banana image generation with model-tier pricing.

Image, video, and audio generation with 600+ models (Flux, SD, Recraft, Grok).

Analyze images with Google Cloud Vision ML. Detect objects, faces, text/OCR, labels, landmarks, logos, safe-search signals, web entities, crop hints, dominant colors, and product-search reference images across image, PDF, and TIFF inputs.

Analyze video content asynchronously with Google Video Intelligence. Detect labels, shots, explicit content, speech, on-screen text, logos, object tracks, scene changes, and searchable metadata for uploaded or cloud-hosted videos.

Generate natural-sounding speech from text or SSML in 50+ languages and 380+ voices. Supports pitch, speed, effects profiles, MP3/WAV/OGG output, long-form synthesis, and Neural2/Studio voice models for production audio.

Convert speech audio to text with Google Speech-to-Text. Supports 125+ languages, short and long audio, streaming-style workflows, speaker diarization, word timestamps, phrase hints, custom classes, profanity filtering, and punctuation.

Analyze text with Google NLP: sentiment, emotion, named entities, entity salience, content categories, moderation categories, syntax, and all-in-one annotation. Supports multilingual text analytics for documents, reviews, tickets, and artic…

Use Google Gemini models for text generation, multimodal image/audio/video understanding, chat, embeddings, code generation, JSON output, function calling, cached context, file search, grounding with Google Search, and fine-tuned models.

Extract structured data from PDFs, scans, and document images using OCR and ML. Handles invoices, receipts, forms, contracts, tax documents, IDs, and custom schemas, returning text, tables, entities, fields, and confidence signals.

Transcribe audio files with Qwen ASR through Alibaba Cloud Model Studio's asynchronous speech recognition API.

Search and retrieve data with Alibaba Cloud PAI Model Gallery, including retrieve the ModelGallery model list, for search and discovery workflows.

Alibaba Cloud Model Studio's OpenAI-compatible API for Qwen chat, reasoning, and code-generation models.

Run image and video analysis or generation with Alibaba Cloud IVPD, including resize Image, element detection, and style Transfer.

Process speech and audio with Alibaba Cloud Intelligent Speech Interaction, including short-form voice command recognition, conversational speech input, and voice search, for transcription and voice workflows.

Run image and video analysis or generation with Alibaba Cloud Image Segmentation, including sky Replacement, facial Feature Segmentation, and fine-Grained Mask Segmentation.

Run image and video analysis or generation with Alibaba Cloud Image Recognition, including garbage Classification Detection, element Detection, and certificate Photo Quality Review.

Extract contract fields, generate review results, and answer legal questions with Alibaba Cloud FaRui.

Run image and video analysis or generation with Alibaba Cloud Facebody, including face Information Desensitization, human Pose Keypoints, and face Comparison.

Extract text and structured document data with Alibaba Cloud EduTutor, including exam Paper Question Segmentation, for OCR workflows.

Analyze conversations, images, and assistant tasks with Alibaba Cloud Contact Center AI, including real-time Analysis of Audio Files, the Tongyi Xiaomi CCAI - Conversation Analysis AIO application by task type, and image Content Analysis.

Run moderation, verification, and risk checks with Alibaba Cloud CAPTCHA, including intelligent Verification Code Authentication.

Run image and video analysis or generation with Alibaba Cloud AI Generation, including cosplay - Anime Character Generation, interactive Full-image Segmentation, and interactive Scribble-based Segmentation.

Uncensored, privacy-preserving LLM and image API (OpenAI-compatible). x402 wallet top-up model with 10 USDC prepay.

Pay-per-query Claude (Haiku/Sonnet/Opus) plus 8 packaged tools. No subscription, no account.

Pay-per-call DeepSeek chat API via x402 micropayments. No signup, no API keys -- just a USDC wallet on Base. Full chat completions endpoint compatible with any x402 client.

Answer Engine Optimization audit — scores any website visibility in AI assistants (ChatGPT, Claude, Perplexity, Gemini). Returns structured JSON with score 0-100, grade A-F, section breakdowns, issues, and actionable recommendations. Checks…

External governance layer for agent systems by Levels Of Self. Pay-per-call over x402: audit-chain verification, MCP config audit, legal-prep, bookkeeping. USDC on Base, no signup.

AI music and video generation agent selling fixed-price, pay-per-request media jobs via x402 and Virtuals ACP on Base. Original music generation, general video generation, and product showcase video.

Ask Eva: the autonomous research desk for crypto, DeFi and global markets, from the Everything.inc ecosystem. Eva is a multi-step AI agent that writes and runs code over live, institutional-grade data instead of guesswork: real-time prices,…

Text or image to production-ready 3D models (GLB, PBR textures, up to 1M triangles) in one API call. Built for AI agents: x402 USDC on Solana, settle-upfront with automatic refund on failure. Also auto-rigs models (skeleton + skinning) and …

Agent Commerce Protocol (ACP) x402 endpoint by Virtuals Protocol on Base. GET /acp-budget serves ACP budget data behind an HTTP 402 paywall.