Run Claude Code with 200 free requests via NVIDIA NIM — 60-second setup

If you've been hitting Anthropic's $5/month minimum or just want a sandbox for Claude Code without funding a balance, point it at InferAll instead. Two env vars, one CLI command, 200 free calls before any signup nag. ## Setup ```bash export ANTHROPIC_BASE_URL=https://api.inferall.ai export ANTHROPIC_API_KEY=your_inferall_key claude ``` That's it. The InferAll key is an `ifu_...` token you create at [inferall.ai/keys](https://inferall.ai/keys). No card needed to sign up; the trial is 200 requests against NVIDIA NIM open-source models. ## What actually runs By default, Claude Code's requests get routed to: - `nvidia/nemotron-3-super-120b-a12b` — 120B parameter frontier-class model. Closest free analogue to Claude Opus / GPT-4 on most reasoning benchmarks. - `meta/llama-3.1-70b-instruct` — 70B mid-class. Closest to Sonnet 4.6 for daily-driver coding tasks. - `meta/llama-3.1-8b-instruct` — 8B fast-class. Closest to Haiku / GPT-4o-mini for high-volume short queries. All three are `$0 input / $0 output`. The trial cap is 200 successful calls across all three. ## If you want actual Claude The same `ANTHROPIC_BASE_URL` config also accepts a `model="anthropic/claude-sonnet-4-6"` style prefix to force the real Claude provider through InferAll's zero-markup billing — same provider rate, just one API key instead of separate Anthropic + Anthropic-via-bedrock + Anthropic-via-vertex keys. After the 200 trial calls, the $5 starter pack unlocks ongoing free NIM use AND becomes spendable balance for premium routing. ## What's the catch? Honestly nothing for typical Claude Code usage. The free NIM models are good enough for most code generation, refactoring, and chat tasks. Where the catch shows up: - **Tool calls / agentic flows**: the NIM models support OpenAI-style function calls; Anthropic's tool-use schema is auto-translated. Mostly works, occasional edge cases. - **Long-context (128K+)**: Llama 3.1 supports 128K but quality at the tail varies. For >50K context, Claude Sonnet via the prefix usually wins. - **Vision**: NIM doesn't currently expose multimodal models; vision routes to Anthropic/Gemini directly (charged from balance). ## Why this exists InferAll is an AI gateway — one base URL, one API key, every major provider routed underneath at the provider's published rate with zero markup. The free NIM tier exists because NVIDIA gives us free inference on the open-source catalog and it's cheaper to pass that through than to gatekeep it. The $5 starter pack is the conversion event for users who want premium-on-demand; everyone else stays in the free tier indefinitely and we're fine with that. If you want to read the [docs](https://inferall.ai/docs), they cover the OpenAI SDK route, the raw REST endpoint, and the trial-status headers SDKs can pick up to surface usage inline.