Image generation API, text to image: images from 36 models (Nano Banana, GPT Image, Flux, Seedream, Recraft, Qwen and more; list at GET /v1/images/models), no API key needed. Call it when an agent needs an image made from a text prompt. OpenAI images request format. Served by SingIt.

Complete attributed investment case, claims, risks, conditions, revisions, measured outcomes and permitted evidence. Paragraph and support selectors provide requested depth. Full input/output contracts: https://beliefstate.ai/v1/research/get_belief

Base mainnet transaction finality check: given a tx hash, classifies it as not_found, pending, unsafe (sequencer-confirmed only), safe (L1-attested), or finalized (L1-finalized) -- read directly from the node's own safe/finalized block tags, not a modeled reorg probability. Input: tx hash. Output: JSON.

Text to speech (TTS) API: text read aloud by 11 voice models (Kokoro, ElevenLabs, Gemini, xAI, MiniMax and more; list at GET /v1/audio/models), no API key needed. Call it when an agent needs spoken audio made from text. OpenAI speech request format, mp3 by default. Served by SingIt.

Domain lookup over RDAP (the successor of WHOIS) and DNS: registered or available, registrar, creation, update and expiry dates, days until expiry, age, status codes, DNSSEC, name servers, plus whether it resolves (IPv4/IPv6) and its mail servers. Any TLD with an RDAP service; others are inferred from DNS.

DeFi pool APY (total, base, reward), TVL, 7/30-day APY change and impermanent-loss estimate (il7d, ilRisk) from DefiLlama. Filter by project, chain and symbol; returns up to 20 pools by TVL. APY is DefiLlama's estimate, not a guaranteed return.

Linkrra Synap: Qwen3 30B A3B Instruct 2507 (model "qwen/qwen3-30b-a3b") - general, high-throughput. OpenAI-compatible chat completions for this one model, 64K context, served only on dedicated GPUs we operate, 4-bit AWQ. Pay per call in USDC on Base via x402 - no account, no API key. From $0.001 per call, usage-priced per token above that (GET /v1/models). Streaming supported. Product site: https://synap.linkrra.com

USDC yield, best lending rates now ($0.002): supply and borrow APY of USDC on every Aave v3, Spark and Compound v3 market (Ethereum, Base, Arbitrum, Optimism, Polygon), utilization and size, highest APY first, read on chain every 10 minutes. Use to park idle USDC.

GPT-5.6 Sol (OpenAI) LLM chat completions, OpenAI-compatible, with no API key or account. One of OpenAI's GPT-5.6 model variants: general-purpose chat, coding and reasoning. Call it when an agent needs GPT-5.6 Sol specifically. Served by SingIt.

Anthropic Messages API for LLM inference (what Claude Code and the Anthropic SDK speak), any model listed at GET /v1/models, Claude included. Call it when an agent needs a Claude-format answer with no API key or account, or wants to try another model the same way. Served by SingIt.

LLM chat completions API, OpenAI-compatible: one endpoint for 100+ models (Claude, GPT, Gemini, Grok, DeepSeek, Qwen), no API key or account, pay per call in USDC. Call it when an agent needs an LLM answer, a second opinion, or a model it has no key for. SingIt Ask; models at GET /v1/models.

GPT-5.6 Terra (OpenAI) LLM chat completions, OpenAI-compatible, with no API key or account. Another GPT-5.6 variant from OpenAI: general-purpose chat, coding and reasoning. Call it when an agent needs GPT-5.6 Terra specifically. Served by SingIt.

OpenAI Responses API for LLM inference, any model listed at GET /v1/models (GPT, Claude, Gemini, DeepSeek and more). Call it when an agent needs an answer in OpenAI's Responses format with no API key or account. Served by SingIt.

Web answers with citations from Exa: a cited answer to a question. Call it when an agent needs a sourced answer rather than raw search results, with no API key or account. Served by SingIt.

Claude Sonnet 5.5 (Anthropic) LLM chat completions, OpenAI-compatible, with no API key or account. Anthropic's mid-sized Claude model: everyday coding, writing and analysis at a lower cost than Opus. Call it when an agent needs Claude Sonnet 5.5 specifically. Served by SingIt.

Claude Haiku 5.5 (Anthropic) LLM chat completions, OpenAI-compatible, with no API key or account. Anthropic's smallest, fastest Claude model: quick, low-latency answers. Call it when an agent needs Claude Haiku 5.5 specifically. Served by SingIt.

Web search API from Exa: up to 25 results with optional page contents. Call it when an agent needs current web results, with no API key or account. Served by SingIt.

Kimi K3 (Moonshot AI) LLM chat completions, OpenAI-compatible, with no API key or account. Moonshot AI's Kimi K3 model: long-context chat and reasoning. Call it when an agent needs Kimi K3 specifically. Served by SingIt.

GLM 5.3 (Zhipu AI) LLM chat completions, OpenAI-compatible, with no API key or account. Zhipu AI's GLM 5.3 model: general-purpose chat and reasoning. Call it when an agent needs GLM 5.3 specifically. Served by SingIt.

GPT-5.5 (OpenAI) LLM chat completions, OpenAI-compatible, with no API key or account. OpenAI's GPT-5.5 model: general-purpose chat, coding and reasoning. Call it when an agent needs GPT-5.5 specifically. Served by SingIt.