Model comparison
Amazon Nova Lite vs DeepSeek-V3.2-Exp
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 31.9 on the Noometry Index. Amazon Nova Lite costs 2.8× less per token, which makes it the better buy when DeepSeek-V3.2-Exp's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
DeepSeekSummary
- They share 18 benchmarks with published results for both. Amazon Nova Lite scores higher in 0 categories and DeepSeek-V3.2-Exp in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.2-Exp leads 51.7 to 26.3.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 per million input/output tokens, against $0.26 / $0.38 for DeepSeek-V3.2-Exp.
- Amazon Nova Lite accepts more context: 300K tokens versus 164K.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Lite | DeepSeek-V3.2-Exp | |
|---|---|---|
| Provider | Amazon | DeepSeek |
| Noometry Index | 31.9 | 44.3 |
| Released | 2024-12-03 | 2025-09-29 |
| Weights | Proprietary | Open |
| Context window | 300K | 164K |
| Max output | 10K | 66K |
| Input $ / M tokens | $0.06 | $0.26 |
| Output $ / M tokens | $0.24 | $0.38 |
| Results tracked | 34 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Exp leads
Amazon Nova Lite: 32.5 (#270), DeepSeek-V3.2-Exp: 46.5 (#65)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Coding | 1239 | 1454 |
| SWE-bench Verified (bash only) | — | 70% |
| Aider Polyglot | — | 74.2% |
| LMArena WebDev | — | 1362 |
| SWE-bench Multilingual | — | 59% |
| SciCode | — | 38.9% |
| WeirdML | — | 39.5% |
| LiveBench Coding | 27.5% | — |
| ALE-Bench | 236.25 | — |
Agentic & Tool Use Not comparable
Amazon Nova Lite: —, DeepSeek-V3.2-Exp: 32.7 (#59)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| Terminal-Bench | — | 39.6% |
| APEX-Agents | — | 21.3% |
| Berkeley Function Calling Leaderboard | — | 56.7% |
| TheAgentCompany | — | 42.9% |
| Vending-Bench 2 | — | 1,034 |
Reasoning DeepSeek-V3.2-Exp leads
Amazon Nova Lite: 19.3 (#260), DeepSeek-V3.2-Exp: 22.1 (#208)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1434 |
| ARC-AGI-2 | — | 4% |
| Kagi LLM Benchmark | — | 52.2% |
| NYT Connections (extended) | — | 36.7% |
| ARC-AGI-1 | — | 57% |
| CritPt | — | 2.9% |
| Chess Puzzles | — | 14% |
| Thematic Generalization | — | 65% |
| LiveBench Reasoning | 36.7% | — |
| DTBench | — | 87.7% |
| LiveBench Data Analysis | 37.2% | — |
| LMCA | — | 29.1% |
| Epoch Capabilities Index | — | 146.27 |
| LiveBench | 36.4% | — |
Math DeepSeek-V3.2-Exp leads
Amazon Nova Lite: 27.8 (#247), DeepSeek-V3.2-Exp: 41.7 (#87)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Math | 1227 | 1435 |
| MathArena Final-Answer Competitions | — | 57.7% |
| OTIS Mock AIME 2024-2025 | — | 87.8% |
| ProofBench | — | 8% |
| Omni-MATH | 23.3% | — |
| LiveBench Math | 36.7% | — |
| FrontierMath (Feb 2025 set) | — | 22.1% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge DeepSeek-V3.2-Exp leads
Amazon Nova Lite: 26.3 (#259), DeepSeek-V3.2-Exp: 51.7 (#66)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| Vectara Hallucination Rate | 6.1% | 5.3% |
| LMArena Expert | 1201 | 1436 |
| GPQA Diamond | — | 83.4% |
| Humanity's Last Exam | 3.6% | — |
| MMLU-Pro | 60% | — |
| GPQA (HELM) | 39.7% | — |
| MMLU | 77% | — |
Multimodal Not comparable
Amazon Nova Lite: 25.5 (#123), DeepSeek-V3.2-Exp: —
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Vision | 990 | — |
Multilingual DeepSeek-V3.2-Exp leads
Amazon Nova Lite: 38.0 (#232), DeepSeek-V3.2-Exp: 52.2 (#90)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Non-English | 1208 | 1409 |
| LMArena Chinese | 1225 | 1461 |
| LMArena French | 1237 | 1433 |
| LMArena German | 1229 | 1440 |
| LMArena Japanese | 1153 | 1374 |
| LMArena Korean | 1153 | 1371 |
| LMArena Russian | 1216 | 1424 |
| LMArena Spanish | 1226 | 1440 |
Instruction Following DeepSeek-V3.2-Exp leads
Amazon Nova Lite: 59.3 (#255), DeepSeek-V3.2-Exp: 74.5 (#93)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Instruction Following | 1205 | 1413 |
| LiveBench Instruction Following | 54.1% | — |
| IFEval | 77.6% | — |
Long Context DeepSeek-V3.2-Exp leads
Amazon Nova Lite: 37.4 (#217), DeepSeek-V3.2-Exp: 47.6 (#16)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Longer Query | 1234 | 1428 |
| Fiction.LiveBench | — | 83.3% |
| CL-bench | — | 13.2% |
| CL-bench Life | — | 9.5% |
Writing & Preference DeepSeek-V3.2-Exp leads
Amazon Nova Lite: 42.1 (#237), DeepSeek-V3.2-Exp: 62.4 (#77)
| Benchmark | Amazon Nova Lite | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Text | 1229 | 1425 |
| LMArena Creative Writing | 1197 | 1403 |
| LMArena Multi-Turn | 1199 | 1427 |
| EQ-Bench Creative Writing | — | 1515 |
| WildBench | 75% | — |
| LiveBench Language | 25.9% | — |
Frequently asked questions
Is Amazon Nova Lite better than DeepSeek-V3.2-Exp?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 31.9 on the Noometry Index. Amazon Nova Lite costs 2.8× less per token, which makes it the better buy when DeepSeek-V3.2-Exp's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Lite or DeepSeek-V3.2-Exp?
Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; DeepSeek-V3.2-Exp lists at $0.26 and $0.38.
Is Amazon Nova Lite or DeepSeek-V3.2-Exp better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.5 versus 32.5 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Lite does, with 300K tokens against 164K.
How many benchmarks do Amazon Nova Lite and DeepSeek-V3.2-Exp share?
18 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and DeepSeek-V3.2-Exp has 49.