Nova 2 Lite
Amazon · released
Amazon's current workhorse: a low-latency multimodal model that reads text, images, video and documents, with optional step-by-step reasoning and a 1M-token context. There is no Nova 3 yet, and the stronger Nova 2 Pro never left limited preview.
Best for: AWS-native workloads · High-volume document processing
- Intelligence Index
- —
- Not yet evaluated
- Price per 1M tokens
- $0.33 in $2.75 out
- Blended $0.94 at 3:1
- Context window
- 1M
- Up to 66K output
- Output speed
- 177 tokens/s
- Median, measured by Artificial Analysis
Capability profile
One 0–100 score per area, placing this model between the weakest and strongest tracked model across that area's benchmarks.
- OverallNot enough data
Based on 1 of 3 benchmarks
- CodingNot enough data
Based on 0 of 5 benchmarks
- Agents & knowledge work0 · #40 of 40
Based on 1 of 6 benchmarks
- Reasoning & science2 · #38 of 39
Based on 2 of 2 benchmarks
- Knowledge & reliability9 · #38 of 39
Based on 2 of 2 benchmarks
- Vision23 · #28 of 31
Based on 1 of 2 benchmarks
Benchmark results
All third-party results on record. Rank is among the tracked models that have been tested on that benchmark.
Overall
- Artificial Analysis Intelligence IndexNot tested
- Vals IndexNot tested
- Arena · Text1335 ±6
#33 of 33· 12,392 votes· Arena
Coding
- Terminal-Bench 4.0Not tested
- SciCodeNot tested
- Arena · WebDevNot tested
- Vibe Code BenchNot tested
- Code MigrationNot tested
Agents & knowledge work
- GDPval-AA0.0
#38 of 39· High effort· Artificial Analysis
- Arena · AgentNot tested
- Finance Agent v2Not tested
- Excel Modeling BenchmarkNot tested
- Legal Research BenchNot tested
- Tax Agent BenchNot tested
Reasoning & science
- Humanity's Last Exam11.6%
#38 of 39· High effort· Artificial Analysis
- CritPt0.3%
#35 of 39· High effort· Artificial Analysis
- ARC-AGI-2saturatingNot tested
- ARC-AGI-3Not tested
- GPQA Diamondsaturating81.1%
#23 of 27· High effort· Artificial Analysis
Knowledge & reliability
#39 of 39· High effort· Artificial Analysis
- AA-LCR60.3%
#36 of 39· High effort· Artificial Analysis
Vision
- MMMU-Pro63.8%
#25 of 27· High effort· Artificial Analysis
- Arena · VisionNot tested
Reported by Amazon
Numbers the lab published at launch, run on its own harness and settings. Useful context, but not independently verified and never used in the rankings on this site.
- SWE-bench Verified53.6%
- GPQA Diamond79.6%
Specifications
- Lab
- Amazon
- Released
- Dec 2, 2025
- Availability
- Generally available
- Context window
- 1,000,000 tokens
- Max output
- 65,536 tokens
- Accepts
- Text, Images, Video, PDFs
- Knowledge cutoff
- Oct 2025
- Weights
- Closed
- Reasoning control
- Off by default · effort low to high
- API model ID
amazon.nova-2-lite-v1:0
Pricing
- Input / 1M tokens
- $0.33
- Cached input / 1M
- $0.08
- Output / 1M tokens
- $2.75
- Blended (3:1)
- $0.94
US on-demand price on Amazon Bedrock. Global cross-Region inference is $0.30 in / $2.50 out.
Worth knowing
- Artificial Analysis only estimates its Intelligence Index, so it has no overall score here.
Official sources
23 benchmarks tracked. See methodology and sources.