Find the Best AI Tools for Any Task
Explore 7,000+ curated AI products across 50 categories. Learn features, compare prices, and upvote your favorites.
Search Results (19 found)
Jev for Agents is a free, independent directory of AI agent builds that use Jev, the System One model from TypeSafe AI. It lists 350+ projects, each linked to source code, demo video or original post, grouped into nine architecture patterns.
jsonl.co is a free, local-first JSONL viewer and editor. It streams .jsonl, .ndjson and .json files of 1GB+ straight in your browser with syntax highlighting, real-time validation and a side-by-side raw/parsed view. Nothing is uploaded and no account is needed. Includes free converters, a validator and guides.
RouteMux is an AI gateway (AI router) for developers and teams. Change one base_url and keep using the OpenAI or Anthropic SDK you already know to reach models from OpenAI, Anthropic, Google, xAI, DeepSeek and more — text, image, video and audio. Reliability comes first: multi-layer redundancy with health-aware routing and automatic failover keeps production workloads up. Billing is a prepaid wallet with one bill across all models — no subscription, no seats — and only successful requests are charged. Unused purchased balance can be refunded at any time.
Access leading AI video, image, and language models through one API key, with model endpoints and reusable generation workflows.
MCPALL is an MCP server directory with original guides, install commands, and client setup for Claude, Cursor, VS Code and Windsurf.
Mac VM manager with a built-in MCP server: let Claude Code or any MCP client start, snapshot and run commands in sandboxed VMs on Apple Silicon. Runs Windows 11 ARM, macOS ARM and Linux ARM VMs. 7-day free trial, then $49 one-time.
APIKO AI API is an AI API gateway for developers. Compare models from OpenAI, Anthropic, Google, DeepSeek, Moonshot AI and others by price and availability, then call them all through one OpenAI-compatible endpoint with one key. Pay as you go, no subscription, and failed calls cost nothing.
HOL Guard runs on your own machine and checks what AI coding agents try to do (shell commands, secret reads, MCP tool connections, package installs) against your policy, then allows the action, pauses it for your decision, or blocks it. Apache-2.0 open-source core plus an optional paid hosted Guard Cloud for teams (Freemium). Install with pipx install hol-guard && hol-guard init, or use the Azure-signed Windows desktop MSI. Not a gateway or network firewall; it does not claim to stop every prompt injection, and a scan is not a safety guarantee. Made by HOL.
TryAPI offers 81+ production-ready APIs with pay-as-you-go pricing, 99.9% uptime, and sub-200ms response times. Includes currency exchange, IP geolocation, weather, UUID generation, and X/Twitter data APIs. Access a suite of production-ready APIs. Pay only for what you use. Scale without limits. Lightning-fast P99 latency under 200ms, enterprise-grade security with API key encryption and TLS 1.3, simple usage-based pricing with no upfront costs.
Codex Reset Radar is a web dashboard that helps developers track Codex usage limits and uncertain reset windows. It shows the current usage-limit and reset status, keeps 26 weeks of reset history so past patterns stay visible, and surfaces public reset announcements from Tibo beside the measured data. Predictions are presented as evidence-based likelihoods rather than exact countdowns, and stale or missing signals are called out instead of being smoothed over, so a developer can decide whether to continue a heavy AI coding session, pause, or move to work that does not depend on the next reset window. It runs in the browser, is free to read, and requires no installation or account.
JevModel is an independent, bilingual guide and browser workbench for TypeSafe's Jev decision model, aimed at developers and product teams who need a judgment their code can act on. Instead of free-form prose, Jev returns a typed answer: one choice from a set you define, a score on a scale, or the probability that a statement is true. The Jev Workbench ships ten editable workflow presets covering support routing, lead scoring, agent action gating, incident triage, release readiness and more; edit the state and questions and run a real decision in the browser. Each account gets five free runs per UTC day plus a one-time 100K token grant. Thirty-plus guides cover the API, question types, access paths, comparisons with LLM prompts and rules engines, and evaluation.
Jev Model is a System One decision model for developers. It accepts shared state as text, JSON objects, or arrays of text and answers up to eight Choice, Score, or Noul questions in parallel, returning structured outputs with probabilities that applications can use for classification, routing, scoring, guardrails, and human-review workflows. The website provides a free unlimited online playground, documentation, API access, and paid access plans.