Happy Horse 1.0 is now on ModelsLab

Try Now
Skip to main content
Available now on ModelsLab · Language Model

Amazon: Nova Premier 1.0Premier Reasoning Power

Try Amazon: Nova Premier 1.0API Documentation

Master Complex Tasks

1M Context

Process Long Inputs

Handles million-token contexts for large codebases, 400+ page documents, or 90-minute videos.

Multimodal Input

Text Image Video

Analyzes text, images, videos for RAG, function calling, and agentic workflows.

Distillation Teacher

Build Custom Models

Distills accurate, low-latency variants of Nova Pro, Lite, Micro via Amazon Bedrock.

Examples

See what Amazon: Nova Premier 1.0 can create

Copy any prompt below and try it yourself in the playground.

Codebase Analysis

“Analyze this 500k token codebase: identify bugs, suggest refactors, and generate unit tests for the main module. Input: [paste codebase here]. Output optimized Python code.”

Document RAG

“Using RAG on these 400-page financial reports [attach PDFs], answer: What are the quarterly revenue trends and risks? Provide chart summaries and executive insights.”

Video Workflow

“Process this 90-minute product demo video [video URL]: extract key features, generate sales script, and create API calls for CRM integration.”

Agentic Planning

“Plan multistep agentic task: Research market data via web tools, call pricing API, generate report on competitors for electric vehicles.”

For Developers

A few lines of code.
Reasoning. One API call.

ModelsLab handles the infrastructure: fast inference, auto-scaling, and a developer-friendly API. No GPU management needed.

  • Serverless: scales to zero, scales to millions
  • Pay per token, no minimums
  • Python and JavaScript SDKs, plus REST API
API Documentation 
import requests


response = requests.post(
    "https://modelslab.com/api/v7/llm/chat/completions",
    json={
  "key": "YOUR_API_KEY",
  "prompt": "",
  "model_id": ""
}
)
print(response.json())

FAQ

Common questions about Amazon: Nova Premier 1.0

Read the docs

Amazon: Nova Premier 1.0 is the most capable multimodal LLM in the Nova 1.0 family for complex reasoning tasks. It processes text, images, videos with 1M token context. Use it as a teacher for distilling custom models.

Scores 87.4% on MMLU, 82% on Math500, 86.3% on SimpleQA with RAG. Excels in BFCL function calling and SWE-bench for agentic coding. Outperforms peers in half of intelligence-tier benchmarks.

Amazon: Nova Premier 1.0 supports 1,000,000 tokens. Enables analysis of large datasets like codebases or long videos. Max output is 32,000 tokens.

Amazon: Nova Premier 1.0 from late 2024 suits complex tasks via Amazon: Nova Premier 1.0 API. Newer Nova 2 models exist for some workloads. Consider distillation for cost-effective use.

Ideal for RAG systems, agentic apps, video analysis, UI automation. Handles multistep workflows like code generation and document processing. Available in select AWS regions.

Access via Amazon Bedrock API only; proprietary model. Supports text generation with multimodal inputs. Check AWS regions like US East for availability.

Ready to create?

Start generating with Amazon: Nova Premier 1.0 on ModelsLab.

Try Amazon: Nova Premier 1.0API Documentation