Connect your tools to InferAll

Anthropic's terminal-native coding agent. Set ANTHROPIC_BASE_URL to InferAll and ANTHROPIC_MODEL to the model you want; Claude Code sends requests to that model through the Anthropic endpoint.

export ANTHROPIC_BASE_URL=https://api.inferall.ai
export ANTHROPIC_API_KEY=ifu_your_key_here
export ANTHROPIC_MODEL=nvidia/nemotron-3-ultra-550b-a55b

VS Code extension for agentic coding. Open Cline's settings (gear icon), choose the OpenAI Compatible API provider, and fill in these fields.

API Provider: OpenAI Compatible
Base URL:     https://api.inferall.ai/v1
API Key:      ifu_your_key_here
Model ID:     nvidia/nemotron-3-ultra-550b-a55b

Terminal-based AI coding agent. Add InferAll as a custom provider in opencode.json and list each model you want to use; OpenCode rejects models it has not been told about. Set INFERALL_API_KEY in your shell, then run with --model inferall/<model id>.

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "inferall": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "InferAll",
      "options": {
        "baseURL": "https://api.inferall.ai/v1",
        "apiKey": "{env:INFERALL_API_KEY}",
        "headers": { "X-Inferall-Routing": "exact" }
      },
      "models": {
        "nvidia/nemotron-3.5-lightning-30b-a3b": {
          "name": "Nemotron 3.5 Lightning",
          "tool_call": true,
          "limit": { "context": 131072, "output": 1024 },
          "options": { "reasoningEffort": "none" }
        }
      }
    }
  }
}

Terminal coding agent (@earendil-works/pi-coding-agent). Add InferAll as a provider in models.json in Pi's agent directory, set INFERALL_API_KEY, then run with --provider inferall --model <model id>.

{
  "providers": {
    "inferall": {
      "baseUrl": "https://api.inferall.ai/v1",
      "api": "openai-completions",
      "apiKey": "$INFERALL_API_KEY",
      "headers": { "X-Inferall-Routing": "exact" },
      "models": [
        { "id": "nvidia/nemotron-3-ultra-550b-a55b", "maxTokens": 1024 }
      ]
    }
  }
}

AI code editor. In Settings > Models > API Keys, turn on and fill in the OpenAI API Key and Override OpenAI Base URL fields, add the exact model ID as a custom model, restart Cursor, then select the model in a new chat. Bring-your-own-key covers chat; Tab completion keeps using Cursor's own models.

OpenAI API Key:             ifu_your_key_here   (Use OpenAI API Key: ON)
Override OpenAI Base URL:   https://api.inferall.ai/v1   (ON)
Custom model:               nvidia/nemotron-3-ultra-550b-a55b

Open-source coding assistant for VS Code and JetBrains. Add InferAll as an OpenAI-provider model in Continue's config.yaml.

name: InferAll
version: 1.0.0
schema: v1
models:
  - name: InferAll Nemotron
    provider: openai
    model: nvidia/nemotron-3-ultra-550b-a55b
    apiBase: https://api.inferall.ai/v1
    apiKey: ifu_your_key_here

Terminal-based AI pair programming. Set OPENAI_API_BASE and OPENAI_API_KEY, then pass the model with the openai/ prefix.

export OPENAI_API_BASE=https://api.inferall.ai/v1
export OPENAI_API_KEY=ifu_your_key_here
aider --model openai/nvidia/nemotron-3-ultra-550b-a55b

Point ChatOpenAI at InferAll with base_url and api_key parameters.

from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
    model="nvidia/nemotron-3-ultra-550b-a55b",
    base_url="https://api.inferall.ai/v1",
    api_key="ifu_your_key_here",
)

Use the OpenAILike wrapper, which accepts third-party model IDs, with a custom API base.

# pip install llama-index-llms-openai-like
from llama_index.llms.openai_like import OpenAILike
from llama_index.core import Settings
Settings.llm = OpenAILike(
    model="nvidia/nemotron-3-ultra-550b-a55b",
    api_base="https://api.inferall.ai/v1",
    api_key="ifu_your_key_here",
    is_chat_model=True,
)

InferAll is OpenAI-compatible — use the openai/ provider prefix and set api_base.

import litellm
response = litellm.completion(
    model="openai/nvidia/nemotron-3-ultra-550b-a55b",
    messages=[{"role": "user", "content": "Hello"}],
    api_base="https://api.inferall.ai/v1",
    api_key="ifu_your_key_here",
)

One command detects your stack, sets up your key in .env, and rewrites existing OpenAI/Anthropic client constructors to point at InferAll. Run `doctor` anytime to re-check the wiring.

npx @inferall/cli init

Pass base_url and api_key to the OpenAI constructor.

from openai import OpenAI
client = OpenAI(
    base_url="https://api.inferall.ai/v1",
    api_key="ifu_your_key_here",
)
response = client.chat.completions.create(
    model="nvidia/nemotron-3-ultra-550b-a55b",
    messages=[{"role": "user", "content": "Hello"}],
)

Pass baseURL and apiKey to the OpenAI constructor.

import OpenAI from "openai";
const client = new OpenAI({
  baseURL: "https://api.inferall.ai/v1",
  apiKey: process.env.INFERALL_API_KEY,
});

Pass base_url to the Anthropic constructor.

import anthropic
client = anthropic.Anthropic(
    base_url="https://api.inferall.ai",
    api_key="ifu_your_key_here",
)
message = client.messages.create(
    model="nvidia/nemotron-3-ultra-550b-a55b",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)

Set INFERALL_API_KEY in your environment, create an LLM for a custom OpenAI-compatible endpoint, then pass it to your agent.

import os
from crewai import LLM
llm = LLM(
    model="nvidia/nemotron-3-ultra-550b-a55b",
    custom_openai=True,
    base_url="https://api.inferall.ai/v1",
    api_key=os.environ["INFERALL_API_KEY"],
)

Classic reads OPENAI_API_BASE_URL, but its built-in model catalog does not include InferAll's NVIDIA models. Using those models requires adapting Classic's provider and model configuration. These endpoint settings alone are not a complete setup; AutoGPT Platform has separate configuration.

Requires a custom-model adapter
Classic endpoint variable: OPENAI_API_BASE_URL
Model settings: FAST_LLM and SMART_LLM
Select models supported by both Classic and your provider.

For the Python open-interpreter package, set your key and pass both the API base and the model. The openai/ prefix selects the OpenAI-compatible protocol.

export OPENAI_API_KEY=ifu_your_key_here
interpreter --api_base https://api.inferall.ai/v1 \
  --model openai/nvidia/nemotron-3-ultra-550b-a55b