The AI arena is free today

Open Superagent

K2 Horizon 3.7B vs Nova 2 Lite

K2 Horizon 3.7B and Nova 2 Lite are closely matched at 21.4 and 24.7 on the LLM Stats Score.

Institute of Foundation Models · Amazon · Updated for 2026

Which is better?

K2 Horizon 3.7B and Nova 2 Lite are closely matched on the overall LLM Stats Score at 21.4 and 24.7.

In the 3 individual benchmarks reported for both models, K2 Horizon 3.7B wins 2; this is a narrower head-to-head signal than the composite indexes.

Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.

Choose K2 Horizon 3.7B

  • you value its reported benchmark strengths — it wins 2 of 3 exact shared results
  • you want the most recent training data — it shipped Sep 2026
  • you need open weights you can self-host or fine-tune

Choose Nova 2 Lite

  • you want predictable pricing at $0.30/M input and $2.50/M output

At a glance

The differences that matter most.

Core performance indexes
21.4
#215
24.7
#193
20.2
#217
26.3
#175
6.4
#220
12.1
#177
9.7
#152
4.4
#185
Cost, coverage & limits
Benchmark wins
2 of 3
1 of 3
Input price
— / M
$0.30 / M
Output price
— / M
$2.50 / M
Context window
—
1,000,000

Capability indexes

Additional strengths measured across groups of related public benchmarks

2 shared
Index
K2 Horizon 3.7B
Nova 2 Lite
13.3#245
24.3#131
8.5#163
10.8#142
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

8 reported for K2 Horizon 3.7B · 19 for Nova 2 Lite

3 shared

K2 Horizon 3.7B outperforms in 2 benchmarks (BFCL-V4, SWE-Bench Verified), while Nova 2 Lite is better at 1 benchmark (GPQA).

K2 Horizon 3.7B shows notably better performance in the majority of benchmarks.

Fri Oct 30 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Context Window

Maximum input and output token capacity

Only Nova 2 Lite specifies input context (1,000,000 tokens). Only Nova 2 Lite specifies output context (65,536 tokens).

Institute of Foundation Models
K2 Horizon 3.7B
Input- tokens
Output- tokens
Amazon
Nova 2 Lite
Input1,000,000 tokens
Output65,536 tokens
Fri Oct 30 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Nova 2 Lite supports multimodal inputs, whereas K2 Horizon 3.7B does not.

Nova 2 Lite can handle both text and other forms of data like images, making it suitable for multimodal applications.

K2 Horizon 3.7B

Text
Images
Audio
Video

Nova 2 Lite

Text
Images
Audio
Video

License

Usage and distribution terms

K2 Horizon 3.7B is licensed under Apache 2.0, while Nova 2 Lite uses a proprietary license.

License differences may affect how you can use these models in commercial or open-source projects.

K2 Horizon 3.7B

Apache 2.0

Open weights

Nova 2 Lite

Proprietary

Closed source

Release Timeline

When each model was launched

K2 Horizon 3.7B was released on 2026-09-03, while Nova 2 Lite was released on 2025-12-02.

K2 Horizon 3.7B is 9 months newer than Nova 2 Lite.

K2 Horizon 3.7B

Sep 3, 2026

1 months ago

9mo newer
Nova 2 Lite

Dec 2, 2025

10 months ago

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Outputs Comparison

Notice missing or incorrect data?

Judge for yourself.

Run your own prompts against K2 Horizon 3.7B and Nova 2 Lite side-by-side, then vote on the output you prefer.

K2 Horizon 3.7B
✓ Preferred
Nova 2 Lite
Open in Playground

FAQ

Common questions about K2 Horizon 3.7B vs Nova 2 Lite.

Which is better, K2 Horizon 3.7B or Nova 2 Lite?

K2 Horizon 3.7B and Nova 2 Lite are closely matched on the LLM Stats Score at 21.4 and 24.7. K2 Horizon 3.7B is made by Institute of Foundation Models and Nova 2 Lite is made by Amazon. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does K2 Horizon 3.7B compare to Nova 2 Lite in benchmarks?

K2 Horizon 3.7B scores HMMT Feb 26: 75.2%, GPQA: 68.9%, SWE-Bench Verified: 65.6%, BFCL-V4: 64.6%, Terminal-Bench 2.1: 39.7%. Nova 2 Lite scores AIME 2025: 91.0%, LongCodeBench: 84.0%, ScreenSpot: 83.3%, MMLU-Pro: 80.9%, GPQA: 79.6%.

What are the context window sizes for K2 Horizon 3.7B and Nova 2 Lite?

K2 Horizon 3.7B supports an unknown number of tokens and Nova 2 Lite supports 1.0M tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between K2 Horizon 3.7B and Nova 2 Lite?

Key differences include LLM Stats Score (21.4 vs 24.7), multimodal support (no vs yes), licensing (Apache 2.0 vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes K2 Horizon 3.7B and Nova 2 Lite?

K2 Horizon 3.7B is developed by Institute of Foundation Models and Nova 2 Lite is developed by Amazon.