Model comparison
Amazon Nova Experimental Chat 26 01 10 vs DeepSeek-V3
Amazon Nova Experimental Chat 26 01 10 is the stronger model overall, scoring 42.2 to 39.2 on the Noometry Index.
Last verified . 13 shared benchmarks.
DeepSeekSummary
- They share 13 benchmarks with published results for both. Amazon Nova Experimental Chat 26 01 10 scores higher in 8 categories and DeepSeek-V3 in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where Amazon Nova Experimental Chat 26 01 10 leads 42.7 to 34.0.
- DeepSeek-V3 has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 | |
|---|---|---|
| Provider | Amazon | DeepSeek |
| Noometry Index | 42.2 | 39.2 |
| Released | — | 2024-12-26 |
| Weights | Proprietary | Open |
| Context window | — | 164K |
| Max output | — | 164K |
| Input $ / M tokens | — | $0.24 |
| Output $ / M tokens | — | $0.90 |
| Results tracked | 13 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Amazon Nova Experimental Chat 26 01 10: 42.1 (#97), DeepSeek-V3: 41.9 (#99)
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| LMArena Coding | 1447 | 1368 |
| Aider Polyglot | — | 55.1% |
| SciCode | — | 35.8% |
| WeirdML | — | 36.1% |
| BigCodeBench Instruct | — | 50% |
| LiveBench Coding | — | 70.9% |
| BigCodeBench Complete | — | 62.2% |
| HumanEval+ | — | 86.6% |
| MBPP+ | — | 73% |
Agentic & Tool Use Not comparable
Amazon Nova Experimental Chat 26 01 10: —, DeepSeek-V3: —
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| METR Time Horizons | — | 49.6% |
Reasoning Amazon Nova Experimental Chat 26 01 10 leads
Amazon Nova Experimental Chat 26 01 10: 27.8 (#97), DeepSeek-V3: 20.0 (#226)
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| LMArena Hard Prompts | 1415 | 1365 |
| SimpleBench | — | 27.2% |
| Kagi LLM Benchmark | — | 52.3% |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 65.8% |
| DTBench | — | 64.8% |
| LiveBench Data Analysis | — | 60.9% |
| LMCA | — | 15.5% |
| BIG-Bench Hard | — | 87.5% |
| Epoch Capabilities Index | — | 135.95 |
| ForecastBench | — | 59.1 |
| HellaSwag | — | 88.9% |
| LiveBench | — | 66.9% |
| PIQA | — | 84.7% |
| WinoGrande | — | 85.2% |
Math Amazon Nova Experimental Chat 26 01 10 leads
Amazon Nova Experimental Chat 26 01 10: 35.8 (#136), DeepSeek-V3: 30.4 (#208)
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| LMArena Math | 1404 | 1373 |
| OTIS Mock AIME 2024-2025 | — | 37.8% |
| Omni-MATH | — | 40.3% |
| LiveBench Math | — | 73.5% |
| MATH Level 5 | — | 75.5% |
| FrontierMath (Feb 2025 set) | — | 1.7% |
Knowledge Amazon Nova Experimental Chat 26 01 10 leads
Amazon Nova Experimental Chat 26 01 10: 39.7 (#115), DeepSeek-V3: 37.5 (#147)
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| LMArena Expert | 1445 | 1352 |
| GPQA Diamond | — | 67.6% |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 26.1% |
| Vectara Hallucination Rate | — | 6.1% |
| GPQA (HELM) | — | 53.8% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 87.2% |
| TriviaQA | — | 82.9% |
Multilingual Amazon Nova Experimental Chat 26 01 10 leads
Amazon Nova Experimental Chat 26 01 10: 50.4 (#121), DeepSeek-V3: 48.8 (#138)
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| LMArena Non-English | 1381 | 1358 |
| LMArena Chinese | 1373 | 1390 |
| LMArena Russian | 1393 | 1372 |
| LMArena Spanish | 1390 | 1358 |
| LMArena French | — | 1386 |
| LMArena German | — | 1374 |
| LMArena Japanese | — | 1333 |
| LMArena Korean | — | 1321 |
Instruction Following Too close to call
Amazon Nova Experimental Chat 26 01 10: 73.1 (#121), DeepSeek-V3: 73.0 (#125)
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| LMArena Instruction Following | 1381 | 1345 |
| LiveBench Instruction Following | — | 81.5% |
| IFEval | — | 83.2% |
Long Context Amazon Nova Experimental Chat 26 01 10 leads
Amazon Nova Experimental Chat 26 01 10: 42.7 (#114), DeepSeek-V3: 34.0 (#248)
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| LMArena Longer Query | 1401 | 1353 |
| Fiction.LiveBench | — | 50% |
Writing & Preference Too close to call
Amazon Nova Experimental Chat 26 01 10: 57.6 (#123), DeepSeek-V3: 57.6 (#124)
| Benchmark | Amazon Nova Experimental Chat 26 01 10 | DeepSeek-V3 |
|---|---|---|
| LMArena Text | 1396 | 1375 |
| LMArena Creative Writing | 1331 | 1364 |
| LMArena Multi-Turn | 1380 | 1389 |
| Short-Story Creative Writing | — | 77% |
| EQ-Bench Creative Writing | — | 1472 |
| WildBench | — | 83% |
| LiveBench Language | — | 49.1% |
Frequently asked questions
Is Amazon Nova Experimental Chat 26 01 10 better than DeepSeek-V3?
Amazon Nova Experimental Chat 26 01 10 is the stronger model overall, scoring 42.2 to 39.2 on the Noometry Index.
Is Amazon Nova Experimental Chat 26 01 10 or DeepSeek-V3 better for coding?
They score almost the same on coding (42.1 vs 41.9); test both on your own repository before choosing.
How many benchmarks do Amazon Nova Experimental Chat 26 01 10 and DeepSeek-V3 share?
13 benchmarks have published results for both models. Amazon Nova Experimental Chat 26 01 10 has 13 scored results on Noometry and DeepSeek-V3 has 60.