AI models & inference

Xynaptic AI Insurance — the insurance vertical agent: coverage families, ORIAS broker registry, premium decomposition + an LLM synthesis of your question. POST { question, type?, broker_name?, orias? }. Legal information (France).

Estimate token count and cost for text across LLM models (GPT-4o, Claude, Gemini, Mistral, Llama). Algorithmic — no API calls.

Generate images, priced per image and size

Official 5-star crash-test safety ratings by VIN or year/make/model: overall, frontal, side, side-pole, and rollover star ratings plus driver/passenger breakdowns, rollover probability, and standard safety-feature availability (stability control, collision and lane-departure warning). Includes counts of complaints and recalls on file. For car-shopping, insurance, and fleet-selection agents. Response includes affiliate offers for full VIN history reports.

All vehicle models for a given make from NHTSA vPIC, no key. Returns model id and name for the manufacturer. For AI agents populating make and model selectors and validating vehicle data.

High-speed vector computation microservice to compute dot products, cosine similarities, and ranking for 1536-d float32 embeddings.

Clean structured JSON dataset of the 100 most active autonomous agent repositories, MCP servers, and LLM orchestration frameworks.

Uses the existing Deepgram and local media pipeline with one diarized transcription call. Subtitle cues split at speaker changes, pauses, six seconds or 84 characters. Unknown speakers remain null. Requires a lawful recording; submitted audio leaves this server for Deepgram, with the existing provider opt-out disclosure. Audio, transcript and subtitles are not retained. This synchronous request has no background job or callback.

A comparison table of language models: input, output and cache prices per million tokens, context window, maximum output length, batch discount and a short note on what each is suited to. Filter by vendor, by a maximum price, by the context window you need, or by capability tag. Every row carries the date its prices were last checked, so a stale figure is visible rather than silently wrong.

Convert audio to text with speaker identification — covers up to 10 minutes of audio

GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon

Transcribe audio to text in 100+ languages via Whisper. Send a file URL, get an accurate transcript back in one call. Pay per call with USDC (x402) or credits - archtools.dev

Ask an LLM anything

x402engine — OpenAI's premium reasoning model — extended thinking for complex math, science, and coding problems

x402engine — OpenAI's higher-compute Terra tier for demanding reasoning workloads

x402engine — OpenAI's balanced GPT-5.6 tier for advanced reasoning and agent workflows

x402engine — Google's latest flagship — advanced reasoning, coding, and multimodal with extended thinking

x402engine — Google's Gemini 3.1 Pro Preview on the Gemini Pro compatibility route — advanced reasoning, coding, and multimodal understanding

x402engine — OpenAI's maximum-compute GPT-5.6 tier for the hardest agent tasks

x402engine — OpenAI's reasoning model — excels at math, science, coding, and multi-step problem solving

x402engine — Anthropic's balanced model — excellent coding, reasoning, and creative writing

x402engine — Anthropic's latest Opus model — premium analysis, coding, and long-context autonomous work

x402engine — Anthropic's previous-gen Opus — strong reasoning and creative writing, large context window

x402engine — Anthropic's most powerful model — best for complex analysis, coding, and extended thinking

x402engine — OpenAI's unified flagship model — 1M context, computer use, top-tier across all tasks

x402engine — Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual

x402engine — Anthropic's latest balanced model — excellent coding, reasoning, and creative writing

Transcribe speech from an audio or video file (mp3, m4a, wav, ogg, mp4, webm) up to 15 minutes, using Whisper. Language is detected automatically. Use when you need the words from a voice note, meeting, podcast clip or video. Send {url} or {base64}, optional language (2-letter code). Returns {language, duration, text, segments with start/end times}. Errors: 413 over 15 minutes, 422 not audio.

OpenAI-compatible chat completion. Pay per request in USDC via x402: no signup, no API keys. Price scales with prompt size and max_tokens.

Speech to text (Whisper large v3 turbo). Send a public audio/video URL or base64 audio (max 20 MB); get the transcript, detected language, timed segments and WebVTT subtitles. Also callable as GET with the same fields as query parameters.