Nova 2 Lite Official Specifications, Pricing & Benchmarks
Nova 2 Lite (nova-2-lite-v1) by Amazon AWS, released 2025-12-02, featuring 1,000,000 context window, 65,535 max output tokens, and Text+Image+Video+PDF input.
📋 Nova 2 Lite Official API & System Card Specifications
Verified against official Amazon AWS developer documentation & pricing
1. Identity & Capabilities
API Spec- API Model ID
- nova-2-lite-v1
- Release Date
- 2025-12-02
- Weight License
- Proprietary (Undisclosed)
- Native Reasoning
- ✅ Supported
- Function Calling & JSON Schema
- 部分支持
- Batch API 异步批处理
- ✅ Supported (50% off)
2. Context, Output & Modalities
Limits- Max Context Window
- 1,000,000 tokens (1000K)
- Max Output Tokens
- 65,535 tokens
- Input Modalities
- Text + Image + Video + PDF
- Output Modalities
- Text
- Direct China Access
- 🌐 Global / OpenRouter
- Deployment Mode
- Cloud API only (Proprietary)
3. Official Pricing & Cache
USD / 1M- Input / 1M Tokens
- $ 0.3
- Cached Input / 1M
- $ 0.06 -80%
- Output / 1M Tokens
- $ 2.5
🎛️ Official Sampling & Reasoning Parameters
Recommended generation settings & API parameter constraints
📊 Nova 2 Lite Full Benchmark Scorecard
18 public benchmarks · ranks among 154 models · synced 2026-11-20
💻 Coding 4 items
LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code.
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
Terminal-Bench is a benchmark for testing AI agents in real terminal environments. It evaluates how well agents can handle real-world, end-to-end tasks autonomously, including compiling code, training models, setting up servers, system administration, security tasks, data…
MCP Atlas is a benchmark for evaluating AI models on scaled tool use capabilities, measuring how well models can coordinate and utilize multiple tools across complex multi-step tasks.
🤖 Agents & Tools 4 items
τ²-bench retail domain evaluates conversational AI agents in customer service scenarios within a dual-control environment where both agent and user can interact with tools.
τ²-Bench telecom domain evaluates conversational agents in a dual-control environment modeled as a Dec-POMDP, where both agent and user use tools in shared telecommunications troubleshooting scenarios that test coordination and communication capabilities.
TAU2 airline domain benchmark for evaluating conversational agents in dual-control environments where both AI agents and users interact with tools in airline customer service scenarios.
Berkeley Function Calling Leaderboard V4 (BFCL-V4) evaluates LLMs on their ability to accurately call functions and APIs, including simple, multiple, parallel, and nested function calls across diverse programming scenarios.
📐 Math 2 items
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999.
A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks.
👁️ Multimodal 5 items
ScreenSpot is the first realistic GUI grounding benchmark that encompasses mobile, desktop, and web environments.
QVHighlights is a video moment retrieval benchmark for detecting moments and highlights in videos via natural language queries.
RealKIE-FCC is a key information extraction benchmark drawn from real enterprise documents (FCC filings), part of the RealKIE suite of five novel datasets for enterprise key information extraction.
A more robust multi-discipline multimodal understanding benchmark that enhances MMMU through a three-step process: filtering text-only answerable questions, augmenting candidate options, and introducing vision-only input settings.
OCRBench v2: Enhanced large-scale bilingual benchmark for evaluating Large Multimodal Models on visual text localization and reasoning with 10,000 human-verified question-answering pairs across 8 core OCR capabilities.
🧠 Reasoning & Science 2 items
A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.
MultiChallenge is a realistic multi-turn conversation evaluation benchmark that challenges frontier LLMs across four key categories: instruction retention (maintaining instructions throughout conversations), inference memory (recalling and connecting details from previous…
📚 Knowledge & General 1 items
Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).
Closest Self-Hostable Open-Weight Alternatives to Nova 2 Lite
Open-weight models with full HuggingFace weights for private GPU / Mac deployment
Frequently Asked Questions
Is Nova 2 Lite open-source or proprietary?
Nova 2 Lite (nova-2-lite-v1) is a proprietary commercial model by Amazon AWS. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.
What are the context limit and API pricing of Nova 2 Lite?
Nova 2 Lite supports 1000K input context and 65,535 max output tokens, priced at $0.3/1M input ($0.06/1M cached) and $2.5/1M output.