DRUX
API — Beta

Build with DRUX

One API call. Multiple AI models. A confidence-weighted answer with an agreement score — ready for your app.

Quick start

curl -X POST https://drux.space/api/v1/search \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "What is the best way to learn TypeScript?",
    "models": [
      "meta-llama/llama-3.2-3b-instruct:free",
      "google/gemma-4-26b-a4b-it:free"
    ]
  }'

Response format

Every response includes per-model answers and a synthesized best answer with a 0–10 agreement score.

{
  "query": "What is the best way to learn TypeScript?",
  "models": [
    {
      "id": "meta-llama/llama-3.2-3b-instruct:free",
      "name": "Llama 3.2 3B",
      "provider": "Meta",
      "status": "done",
      "response": "Start with the official TypeScript handbook...",
      "durationMs": 1823,
      "error": null
    }
  ],
  "synthesis": {
    "content": "The best way to learn TypeScript is...",
    "score": 9,
    "label": "Strong consensus",
    "agreement": ["Start with official docs", "Practice with real projects"],
    "disagreement": []
  }
}

Endpoint

POST/api/v1/search

Request body

querystringrequiredThe question to ask all models
modelsstring[]Array of model IDs (max 5). Defaults to 3 free models.

Authentication

Authorization: Bearer YOUR_API_KEY

Model reference

Pass any of these IDs in the models array.

Free

meta-llama/llama-3.2-3b-instruct:freeLlama 3.2 3B · Meta
Free
google/gemma-4-26b-a4b-it:freeGemma 4 26B · Google
Free
google/gemma-4-31b-it:freeGemma 4 31B · Google
Free
nvidia/nemotron-nano-9b-v2:freeNemotron 9B · NVIDIA
Free
nvidia/nemotron-3-super-120b-a12b:freeNemotron 120B · NVIDIA
Free
openai/gpt-oss-120b:freeGPT OSS 120B · OpenAI
Free
openai/gpt-oss-20b:freeGPT OSS 20B · OpenAI
Free
arcee-ai/trinity-large-thinking:freeTrinity Thinking · Arcee AI
Free
liquid/lfm-2.5-1.2b-thinking:freeLFM Thinking 1.2B · Liquid AI
Free
qwen/qwen3-coder:freeQwen3 Coder · Alibaba
Free

Paid (requires credits)

openai/gpt-5.5GPT-5.5 · OpenAI
8 cr
openai/gpt-5.4GPT-5.4 · OpenAI
5 cr
openai/gpt-oss-120bGPT OSS 120B · OpenAI
3 cr
openai/gpt-oss-20bGPT OSS 20B · OpenAI
1 cr
openai/gpt-4o-miniGPT-4o Mini · OpenAI
1 cr
anthropic/claude-opus-4.7Claude Opus 4.7 · Anthropic
8 cr
anthropic/claude-sonnet-4.6Claude Sonnet 4.6 · Anthropic
3 cr
anthropic/claude-sonnet-4.5Claude Sonnet 4.5 · Anthropic
3 cr
anthropic/claude-haiku-4.5Claude Haiku 4.5 · Anthropic
1 cr
google/gemini-3.5-flashGemini 3.5 Flash · Google
3 cr
google/gemini-2.5-proGemini 2.5 Pro · Google
8 cr
google/gemini-2.5-flashGemini 2.5 Flash · Google
3 cr
deepseek/deepseek-v3.2DeepSeek V3.2 · DeepSeek
3 cr
deepseek/deepseek-chatDeepSeek V3 · DeepSeek
1 cr
mistralai/mistral-large-2512Mistral Large 3 · Mistral
3 cr
mistralai/mistral-small-3.2-24b-instructMistral Small 3.2 · Mistral
1 cr
mistralai/codestral-2508Codestral 2508 · Mistral
3 cr
meta-llama/llama-4-maverickLlama 4 Maverick · Meta
3 cr
meta-llama/llama-4-scoutLlama 4 Scout · Meta
1 cr
qwen/qwen3-maxQwen3 Max · Alibaba
3 cr
qwen/qwen3-coderQwen3 Coder · Alibaba
3 cr
qwen/qwen3-32bQwen3 32B · Alibaba
1 cr
cohere/command-aCommand A · Cohere
3 cr
cohere/command-r-plus-08-2024Command R+ · Cohere
1 cr
amazon/nova-premier-v1Nova Premier · Amazon
5 cr
amazon/nova-pro-v1Nova Pro · Amazon
3 cr
microsoft/phi-4Phi-4 · Microsoft
1 cr
nvidia/nemotron-3-super-120b-a12bNemotron Super 120B · NVIDIA
3 cr