Model APIs and gateways
Model inference, LLM gateways and tools to pick and test models. 7 resources; API, key, license and embedding quoted from each site's own pages, checked Oct 24, 2026.
Zro
You can point an existing coding agent at cheaper open models by changing a base URL, without request bodies being stored or used for training.
API (paid)Commercial OK
Cohesor
You can cut and cap agent model spend by swapping one base URL and key instead of rewriting code for each provider.
API (paid)Commercial OK
LLMTest
A provider outage or a reply that is not valid JSON no longer breaks your feature, because the proxy retries on another model inside the same request, and you see what each AI feature costs.
API (paid)Commercial OK
Nous Research (Nous Portal inference API)
You can call Hermes models through a standard OpenAI client, or download the same weights from Hugging Face and run them yourself.
APIDownloadFree tierCommercial: conditions
Tokenwise
You find out which prompts and models drive your LLM bill by changing one base URL, while your own provider keys pass through and are not stored.
API (paid)Free tierCommercial OK
Opper AI
Switching or pooling model providers becomes a model-string change, and teams that must keep data in the EU get one sub-processor and one DPA instead of a contract per provider.
APIFree tierCommercial OK
RunInfra
You can point Claude Code, Codex, Aider or an existing OpenAI/Anthropic SDK at open models by changing a base URL and key, without renting or tuning GPUs yourself.
API (paid)Free tierCommercial OK
Side by side
| Resource | API | Key | Data download | Commercial use | Embed |
|---|---|---|---|---|---|
| Zro A hosted inference service for open-weight coding models (GLM, DeepSeek, MiniMax) behind OpenAI- and Anthropic-compatible endpoints, with a CLI that launches Claude Code, Codex, OpenCode and similar tools against it, and per-key EU or US region pinning. |
Official API · Paid | Yes | — | Commercial use OK | Embedding unclear |
| Cohesor A hosted gateway that sits between your agents and model providers: it accepts Anthropic- or OpenAI-style requests at one base URL, compresses prompts, sends simple requests to cheaper models, enforces per-user budgets and exposes a governed MCP endpoint for GitHub, Google, Slack and Postgres tools. |
Official API · Paid | Yes | — | Commercial use OK | Embedding unclear |
| LLMTest An OpenAI-compatible LLM proxy that routes your calls to 340+ models and adds per-feature cost tracking, automatic fallback to another model on errors or invalid JSON, model benchmarks with an AI judge, and optional weekly prompt and model optimization, plus an MCP server for your IDE. |
Official API · Paid | Yes | — | Commercial use OK | Embedding unclear |
| Nous Research (Nous Portal inference API) An AI lab that trains the open-weight Hermes models and runs Nous Portal, an OpenAI-compatible inference API for its own Hermes 4 models and a catalog of other labs' models, with keys, credits and subscriptions, plus pay-per-request x402 payments in Solana USDC without an account. |
Official API · Free tier | Yes | Yes (safetensors) | Commercial use with conditions | Embedding unclear |
| Tokenwise A drop-in edge proxy for OpenAI, Anthropic, Gemini, Grok, OpenRouter and other LLM APIs that logs cost, tokens, latency and errors per call and suggests fixes such as model swaps, caching and prompt trims, with rules for fallbacks, A/B splits and budget caps. |
Official API · Paid | Yes | — | Commercial use OK | Embedding unclear |
| Opper AI An EU-hosted LLM gateway that puts 700+ models from 50+ providers behind one key and an OpenAI-compatible endpoint, with per-request fallbacks, tracing, spend caps, model allowlists and data-retention rules. |
Official API · Free tier | Yes | — | Commercial use OK | Embedding unclear |
| RunInfra A hosted API for open-weight models (GLM, DeepSeek, Qwen, Nemotron and others) that speaks both the OpenAI Chat Completions and Anthropic Messages formats at one base URL, billed per token or through monthly coding plans. |
Official API · Paid | Yes | — | Commercial use OK | Embedding unclear |