Nova 2.0 Omni (medium)

Nova 2.0 Omni (medium)

Amazon · Released Nov 26, 2025 Specialist
Intelligence #144 / 571
51.3 Our Score
AA Index #122 / 374
28.0 Artificial Analysis
Input #327 / 571
$0.300 per 1M tokens
Output #405 / 571
$2.50 per 1M tokens
Context
— Not reported

Analysis Summary

Nova 2.0 Omni (medium) is Amazon's multimodal mid-tier reasoning model, released in November 2025. Its intelligence index of 28.0 is supported by a math index of 89.7, GPQA of 0.76, and MMLU-Pro of 0.809. The agentic index of 42.5 is a standout figure for this intelligence tier, and the TAU2 score of 0.804 indicates strong multi-step task reliability. IFBench of 0.662 and LCR of 0.537 reflect above-average instruction following and long-context capability.

For businesses, this model is well-suited to structured content generation, document analysis, and semi-autonomous workflows where instruction precision and agentic reliability matter. The omni designation suggests multimodal input support, which broadens its applicability across content and analysis tasks. Coding capability (15.1) is limited, so software engineering workloads should use a more coding-focused model.

At $0.30 input and $2.50 output per million tokens, the pricing is competitive for its capability profile. Teams running content-heavy or agentic workflows on AWS infrastructure will find Nova 2.0 Omni (medium) a practical and cost-effective choice.

Assessed June 6, 2026

Editorial notes

Nova 2.0 Omni (medium) from Amazon delivers a strong agentic index of 42.5, excellent instruction following, and a high math index, with multimodal capability and competitive pricing for a mid-tier reasoning model.

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence4.5Technical3.5Value6.5Content7
Intelligence 4.5/10
Technical 3.5/10
Content 7/10
Value 6.5/10

How Nova 2.0 Omni (medium) compares

Nova 2.0 Omni (medium) ranks #122 of 374 AI models we track for overall intelligence, #177 of 311 for coding, #109 of 286 for agentic tasks. At $0.30 per million input tokens it is cheaper than 43% of comparable models.

Performance Indices

Source: Artificial Analysis

28 Intelligence Index
15.1 Coding Index
42.5 Agentic Index
89.7 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 76% Graduate-level scientific reasoning
HLE 6.8% Humanity's Last Exam
MMLU Pro 80.9% Multi-task language understanding
AIME 2025 89.7% Competition mathematics (2025)
SciCode 36.2% Scientific computing

Technical

LiveCodeBench 66% Live coding evaluation
TerminalBench Hard 4.5% Agentic terminal tasks
τ²-Bench 80.4% Conversational agent benchmark

Content

IFBench 66.2% Instruction following
LCR 53.7% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does Nova 2.0 Omni (medium) stack up?

Compare side-by-side with other specialist models.

Compare Models

Model Information

ProviderAmazon
Release Date November 26, 2025
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.30 $0.000300
Output $2.50 $0.002500

Leaderboard Categories

Frequently asked questions about Nova 2.0 Omni (medium)

How much does Nova 2.0 Omni (medium) cost?

Nova 2.0 Omni (medium) costs $0.30 per million input tokens and $2.50 per million output tokens.

Is Nova 2.0 Omni (medium) good for coding?

On our coding benchmark index, Nova 2.0 Omni (medium) ranks #177 of 311 models, placing it in the broader range of the field for code generation and debugging.

Who created Nova 2.0 Omni (medium)?

Nova 2.0 Omni (medium) is developed by Amazon and was released on November 26, 2025.