660 tools
Free tier unknownAI models & APIs
Lightning Rod’s current Foresight v4 offering returns probabilistic forecasts with an optional research step. It uses an OpenAI-compatible interface extended with forecasting answer types.
Free to startChatbots & assistants
discode.ai routes chat requests among hosted models based on user preferences. It describes optional multi-model checking and a local sensitive-data assistant that suggests anonymization before sending a prompt.
Free to startAI models & APIs
RunInfra provides hosted open models through OpenAI- and Anthropic-compatible APIs. It offers prepaid token billing and workspace coding plans, with session-aware caching and separately reported cached-token usage.
Free planAI models & APIs
Opper offers a unified model gateway with SDK compatibility, provider fallbacks and optional organization controls. Its Stockholm-hosted platform advertises model allowlists, budgets, routing and configurable trace retention.
Free planSecurity & privacy
Gate sits between agents and model providers to log requests, compress context and optionally screen content. Its Connect desktop app advertises routing existing coding tools, while the gateway supports provider keys or prepaid model usage.
Free planAI models & APIs
Edgee routes coding-agent requests through configurable model policies, compression and usage reporting. It advertises provider fallback and cost attribution by developer, repository and pull request, with team controls on paid plans.
Free to startAI agents & automation
Coasty operates a computer-use arena and evaluation infrastructure with Windows/Linux environments. It advertises blind outcome-graded agent comparisons, private holdout suites and custom benchmarks built from business workflows.
No free planAI models & APIs
Alvoff hosts Whisper transcription, Kokoro speech generation, Qwen embeddings and a small Qwen coding model behind an OpenAI-compatible API. Requests use prepaid, metered credits rather than a subscription.
Free planAI models & APIs
Auriko provides an OpenAI-compatible model gateway with provider routing, fallback, key orchestration and cost analytics. It supports user-supplied or platform keys and configurable objectives such as cost, latency and data policy.
Free tier unknownAI models & APIs
Thinking Machines’ Tinker is a training API for researchers using LoRA adapters on supported models. It abstracts infrastructure while exposing forward/backward gradient computation, optimizer updates, sampling and checkpoint saving for custom supervised or reinforcement-learning loops.
Free tier unknownAI models & APIs
Google’s Gemini 3.6 Flash family announcement describes coding, knowledge work and multimodal agent capabilities, alongside 3.5 Flash-Lite for throughput-focused workloads. Availability differs between the developer API, enterprise tools and consumer applications.
Free tier unknownAI models & APIs
BaseRT is Base Compute’s local inference runtime for Apple Silicon, advertising supported model families and serving for coding-agent clients. The official page provides installation, documentation and repository links plus model-specific benchmark examples.
Free planAI models & APIs
Polygres adds graph, vector, full-text and hybrid retrieval over PostgreSQL for AI context. Developers can host a database or synchronize selected existing data, explore retrieval in a playground and use a CLI or Python SDK.
Free tier unknownAI models & APIs
Gemini Robotics 2 is Google DeepMind’s model family for robot control and embodied reasoning. Its vision-language-action, ER and on-device variants target whole-body movement, dexterity, multi-step planning and collaboration across robot embodiments.
Free planAI models & APIs
Token Harbor exposes several model providers through one OpenAI-compatible API with free routes and paid usage passes. Users choose model IDs, review usage allowances and decide whether calls pause or continue from a paid balance at the pass limit.
Free tier unknownAI models & APIs
ngrok AI Gateway routes SDK requests across frontier providers and privately reachable self-hosted models. It centralizes scoped keys, configured retries/fallbacks and token, latency and error observability through one managed endpoint.
Free planAI models & APIs
Soup CLI inspects training data, chooses post-training methods through rules, prepares configurations and derives evaluations. It supports adapter lifecycle tools, serving/export workflows and evidence gates for model saves, with optional experimental layer streaming.
Free tier unknownAI models & APIs
Prime Intellect combines GPU compute, reinforcement-learning environments, evaluations, hosted training, inference and execution sandboxes. Developers use the Prime CLI to build environments and evaluate or train agents, then access hosted or third-party models through one gateway.
Free planCoding & development
Paritok sits between coding agents and model providers, filtering tool schemas, compressing file/tool output and summarizing older history. Modified segments retain references so agents can retrieve original content when needed.
Free planAI agents & automation
Oqoqo runs repeated agent experiments on sandboxed cloud machines with chosen tasks, treatments and rubrics. It compares agents, models and tool packages, retaining trajectories, errors, files and pass/fail results for iterative interface improvements.
Free planAI models & APIs
Unsloth Desktop runs and trains local models in a desktop workspace with model downloads, chat and no-code media workflows. It exposes an OpenAI-compatible API and connects coding agents to selected local models, with research and code-execution features.
Free planCoding & development
Caveman wraps supported coding agents or runs as SDK middleware to compress eligible tool results and shorten responses. It preserves original context for supported recovery workflows; its private-preview Cloud adds cost traces, evaluations and proposed optimizations.
Free tier unknownAI models & APIs
Gemini 3.7 Flash is a Google model release for coding, agent workflows and document-heavy knowledge work. The official announcement describes Gemini API access and deployment through Google developer and enterprise surfaces, with a dated introductory token tariff.
Free planAI models & APIs
Inferock Bench is a local diagnostic proxy that records metered LLM calls and generates usage, timing and billing-integrity receipts. It separates observed spend, calculated bill-bounded loss, time loss and invoice-check exposure, with explicit coverage states.
Showing 48 of 660 tools
Plan facts carry individual sources and check dates. Unknown values are excluded from required filters.