Nova 2 Lite
Dec 2025
Nova 2 LiteCompare inon
Benchmarks of Nova 2 Lite15
BenchmarkByScoreRank
Knowledge13
Writing1
Legacy · saturated or no longer reported1
No benchmark matches
Rankingconfiguration unstated?
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGPT-5 Mini (2025-08-07)unstated×2-19%$0.25 / $2Context: 1M → 400K
BetterClaude Opus 4.5unstated×2.9+1,076%$5 / $25Context: 1M → 200K
OpenKimi K2 0711unstated×2.2same$0.50 / $2Context: 1M → 128K+1
CheaperQwen3 235B A22B Instruct 2507unstatedOpen+92%-76%$0.09 / $0.55Context: 1M → 262K+1CheaperGPT-5 Nano (2025-08-07)unstated+90%-84%$0.05 / $0.40Context: 1M → 400KModelScoreCostPer taskWhat changes
GPT-5 Mini (2025-08-07)unstated×2-19%$0.25 / $2Context: 1M → 400KQwen3 235B A22B Instruct 2507unstatedOpen+92%-76%$0.09 / $0.55Context: 1M → 262K+1GPT-5 Nano (2025-08-07)unstated+90%-84%$0.05 / $0.40Context: 1M → 400KGPT-4.1 Mini (2025-04-14)+86%-18%$0.40 / $1.60Qwen3 8BunstatedOpen+57%-64%$0.18 / $0.70Context: 1M → 33K+1ModelScoreCostPer taskWhat changes
Claude Opus 4.5unstated×2.9+1,076%$5 / $25Context: 1M → 200KClaude Sonnet 4.5unstated×2.7+606%$3 / $15Claude Haiku 4.5unstated×2.5+135%$1 / $5Context: 1M → 200Ko3 (2025-04-16)unstated×2.3+312%$2 / $8Context: 1M → 200KKimi K2 0711unstatedOpen×2.2same$0.50 / $2Context: 1M → 128K+1ModelScoreCostPer taskWhat changes
Kimi K2 0711unstatedOpen×2.2same$0.50 / $2Context: 1M → 128K+1Qwen3 235B A22B Instruct 2507unstatedOpen+92%-76%$0.09 / $0.55Context: 1M → 262K+1NNanbeige4 3B Thinking 2511Open+90%——Context: 1M → 66K+1Qwen3 32BunstatedOpen+80%+44%$0.70 / $2.80Context: 1M → 33K+1
Command AunstatedOpen+72%+415%$2.50 / $10Context: 1M → 256K+1ModelScoreCostPer taskWhat changes
Nova 2.0 Pro Previewmediumn/a——Context: 1M → 256KNova 2.0 Omnimediumn/a—$0.30 / $2.50Nova 2.0 Litemediumn/a—$0.30 / $2.50Nova Premiernon-thinkingn/a—$2.50 / $12.50Nova Pron/a—$0.80 / $3.20Context: 1M → 300KNova Liten/a—$0.06 / $0.24Context: 1M → 300KNova Micron/a—$0.035 / $0.14Context: 1M → 128K+1Nova Lite 1.0n/a—$0.06 / $0.24Context: 1M → 300KAll Amazon models →Benchmarks
all categoriesChat & Knowledge
Arena Text · Chinese1,364.4#121
Arena Text · English1,384.6#106
Arena Text · French1,380.6#99
Arena Text · German1,342.9#102
Arena Text · instruction following1,334.9#110
Arena Text · Japanese1,273.2#93
Arena Text · Korean1,283.2#99
Arena Text · non-English1,337.1#113
Arena Text · Polish1,327.1#100
Arena Text · Russian1,342.2#112
Arena Text · Spanish1,372.6#101
Arena Text · overall · full cohort1,361.8#110
Agents & Tool Use
BFCL v427.1#29
Writing & Language
Arena Text · creative writing · full cohort1,291.8#124
FAQ
How much does Nova 2 Lite cost?›
$0.30 per 1M input tokens and $2.50 per 1M output tokens.
Is Nova 2 Lite free to use?›
There is no free API; the cheapest paid price is $0.30 per 1M input tokens.
How good is Nova 2 Lite, and what is it best at?›
#106 of 188 in Chat & Knowledge, #29 of 39 in Agents & Tool Use, #124 of 187 in Writing & Language.
Is Nova 2 Lite better than Llama 4 Scout?›
On BFCL v4, Nova 2 Lite scores 27.1 vs 28.1 — lower.
How can I use Nova 2 Lite?›
Through the API of Amazon Bedrock.
What is the context window of Nova 2 Lite?›
1M tokens, with up to 66K output tokens.
When was Nova 2 Lite released? Is there a newer version?›
Released 2025-12-02. It is the current version in its line.
Is Nova 2 Lite open source?›
No, Nova 2 Lite is a closed model available through an API. The closest open-weights option here is Kimi K2 0711.
Can Nova 2 Lite read images?›
It accepts text and images, PDF files, video as input.
What is Nova 2 Lite?›
Nova 2 Lite is a closed language model from Amazon, released Dec 2025.
What are the best alternatives to Nova 2 Lite?›
GPT-5 Mini (2025-08-07) (cheaper), Claude Opus 4.5 (better), Kimi K2 0711 (open weights).