InferAll — Unified AI API for OpenAI, Claude, Gemini & 207+ Models<br>Run Claude Code on freeOSS models. Upgrade whenyou need to.<br>InferAll is an Anthropic-compatible gateway with $0 input/output on 118+ NVIDIA-hosted open-source models (within the free-plan daily request limit), pay-as-you-go access to Claude, GPT-4, and Gemini, and automatic provider failover. Point Claude Code, Cline, Cursor, LangChain or any OpenAI-compatible client at one base URL — one key, one bill.<br>Drop into Claude Code with two env vars — no code changes.<br>Drop-in for Claude Code<br>export ANTHROPIC_API_KEY=ifu_...<br>export ANTHROPIC_BASE_URL=https://api.inferall.ai<br>claudeCopy
Or call the gateway directly with the official SDK — Python (inferall-ai 0.1.0 on PyPI) or TypeScript (@inferall/sdk 0.1.0 on npm).<br>Python SDK — install<br>pip install inferall-aiCopy
Python SDK — usage<br>from inferall import Inferall
ai = Inferall() # reads INFERALL_API_KEY<br>ai.text("Explain quantum computing in two sentences")Copy
Get free API keyHow failover works<br>$5 starter pack unlocks unlimited NVIDIA NIM open-source access (118+ models) — the $5 becomes spendable balance on any premium model. See /pricing.<br>219 models live · 121 free · 14,200 requests / 30d · gateway healthy
Why InferAll<br>Open models at $0 rate<br>118+ open-source models hosted on NVIDIA NIM — Llama 3.1 70B, Mixtral, Nemotron, CodeLlama — bill at our open-model rate ($0 input, $0 output) once your account is activated with the $5 starter pack.
Multi-provider routing<br>One key fronts OpenAI, Anthropic, Google, NVIDIA NIM, Replicate, and Runway. Switch providers with a parameter, not a rewrite.
Drop-in compatibility<br>OpenAI-compatible at /ai/v1/generate. Anthropic-compatible at /v1/messages. Existing SDKs, agents, and editors work unchanged.
One bill<br>Pay-as-you-go on premium providers at their published token price with zero markup. Free OSS, premium when you need it, single invoice.
Resilience<br>Automatic failover across providers<br>Anthropic rate-limits. OpenAI has incidents. Google Gemini quotas reset. Production agents that call a single provider directly inherit every one of those failures.<br>InferAll's gateway watches for 429s, 529s, 5xx responses, and timeouts (30s default). When the primary provider can't serve a request, the same prompt is retried server-side against the next provider that can run an equivalent model — Anthropic to Google, OpenAI to NVIDIA, and so on. Your client code does not change.<br>Context-size-aware routing means a 1M-token prompt is only sent to providers whose models can actually fit it. Cheapest-first selection means you pay the least for the same quality bar.
Failover flow<br># Anthropic returns 429 / 529 / 5xx?<br># InferAll transparently retries against the next provider<br># that can serve the requested model class.
POST https://api.inferall.ai/v1/messages<br>→ Anthropic (rate-limited) [retry]<br>→ Google Gemini 2.5 [200 OK]Copy
Providers<br>OpenAIGPT-4o, DALL-E 3, o1<br>AnthropicClaude Sonnet 4, Opus, Haiku<br>GoogleGemini 2.5, Veo 3, Imagen 4<br>NVIDIALlama 3.1 70B, Mixtral, 118+ free models<br>ReplicateFlux, Stable Diffusion, video<br>RunwayGen-4.5, video generation
Editor integration<br>Use with your favorite editor<br>Anything that speaks the OpenAI or Anthropic wire format works with InferAll. That covers most modern coding agents — including the three below.<br>Claude Code<br>Anthropic's terminal-native coding agent. Point ANTHROPIC_BASE_URL at InferAll and route Claude Code through free OSS models for cheap turns, premium models for hard ones.<br>Setup snippet<br>export ANTHROPIC_API_KEY=ifu_...<br>export ANTHROPIC_BASE_URL=https://api.inferall.ai<br>claudeCopy
Cline<br>The autonomous coding agent for VS Code. Cline supports Anthropic-compatible endpoints — set the base URL in settings and your free OSS allowance covers the chatty inner loop.<br>Setup steps<br># In Cline's Anthropic provider settings:<br># API Key: ifu_...<br># Base URL: https://api.inferall.aiCopy
Cursor<br>Cursor lets you bring your own OpenAI-compatible endpoint. Configure InferAll as a custom provider to use the same key across editors.<br>Setup steps<br># In Cursor → Settings → Models → Custom API:<br># OpenAI API Key: ifu_...<br># OpenAI Base URL: https://api.inferall.ai/v1Copy
Prefer a one-click VS Code install with inference already wired in? Use the dedicated InferAll for VS Code extension — the canonical InferAll-branded experience, with the upstream Cline + base-URL path above still fully supported.<br>Dedicated Cline and Cursor integration guides are coming to the docs.
Solutions<br>Unified AI APIOne key for OpenAI, Claude, Gemini, and Llama<br>LLM API aggregator207+ models across 6 providers, one endpoint<br>AI model gatewayIntelligent routing with automatic provider fallback<br>AI inference API118+ free open-source models plus premium providers<br>Compare AI modelsTest GPT-4o, Claude, Gemini, and Llama side-by-side
How it works<br>01<br>Get your API key<br>Sign up at /keys and create an InferAll API key (prefix ifu_). Activate the key at /billing with the $5 starter pack — that $5...