Model comparison
Amazon Nova Experimental Chat 12 10 vs DeepSeek V4 Pro
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 42.9 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeekSummary
- They share 12 benchmarks with published results for both. Amazon Nova Experimental Chat 12 10 scores higher in 0 categories and DeepSeek V4 Pro in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 29.3.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro | |
|---|---|---|
| Provider | Amazon | DeepSeek |
| Noometry Index | 42.9 | 54.3 |
| Released | — | 2026-04-24 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 393K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1.98 |
| Results tracked | 12 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
Amazon Nova Experimental Chat 12 10: 42.2 (#110), DeepSeek V4 Pro: 52.4 (#34)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Coding | 1430 | 1470 |
| SWE-bench Verified | — | 77.6% |
| FrontierCode | — | 28.6% |
| LMArena WebDev | — | 1582 |
| SciCode | — | 51% |
| WeirdML | — | 66.2% |
| ALE-Bench | — | 1,403 |
Agentic & Tool Use Not comparable
Amazon Nova Experimental Chat 12 10: —, DeepSeek V4 Pro: 32.8 (#58)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| APEX-Agents | — | 47.3% |
| Vending-Bench 2 | — | 3,285 |
Reasoning DeepSeek V4 Pro leads
Amazon Nova Experimental Chat 12 10: 29.3 (#94), DeepSeek V4 Pro: 56.5 (#24)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Hard Prompts | 1427 | 1461 |
| ARC-AGI-2 | — | 61.3% |
| Kagi LLM Benchmark | — | 53.5% |
| NYT Connections (extended) | — | 91.3% |
| ARC-AGI-1 | — | 90.5% |
| CritPt | — | 18% |
| Chess Puzzles | — | 47% |
| Mystery Game Puzzles | — | 43% |
| DTBench | — | 93.9% |
| LMCA | — | 45.5% |
| Surface Evolver Bench | — | 40% |
| Epoch Capabilities Index | — | 155.31 |
| ForecastBench | — | 56.1 |
Math DeepSeek V4 Pro leads
Amazon Nova Experimental Chat 12 10: 38.9 (#124), DeepSeek V4 Pro: 64.8 (#30)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Math | 1419 | 1455 |
| FrontierMath (Tiers 1-3) | — | 64.6% |
| FrontierMath Tier 4 | — | 26.8% |
| MathArena Final-Answer Competitions | — | 76.6% |
| OTIS Mock AIME 2024-2025 | — | 98.6% |
| ProofBench | — | 50% |
Knowledge DeepSeek V4 Pro leads
Amazon Nova Experimental Chat 12 10: 39.2 (#136), DeepSeek V4 Pro: 59.5 (#31)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Expert | 1408 | 1464 |
| GPQA Diamond | — | 91.7% |
| SimpleQA Verified | — | 52.9% |
| Vectara Hallucination Rate | — | 8.6% |
Multilingual DeepSeek V4 Pro leads
Amazon Nova Experimental Chat 12 10: 51.4 (#108), DeepSeek V4 Pro: 54.4 (#45)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Non-English | 1398 | 1439 |
| LMArena Chinese | 1440 | 1486 |
| LMArena Russian | 1395 | 1453 |
| LMArena French | — | 1472 |
| LMArena German | — | 1458 |
| LMArena Japanese | — | 1445 |
| LMArena Korean | — | 1447 |
| LMArena Spanish | — | 1458 |
Instruction Following DeepSeek V4 Pro leads
Amazon Nova Experimental Chat 12 10: 73.2 (#123), DeepSeek V4 Pro: 76.1 (#47)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Instruction Following | 1387 | 1448 |
Long Context DeepSeek V4 Pro leads
Amazon Nova Experimental Chat 12 10: 42.4 (#125), DeepSeek V4 Pro: 45.0 (#51)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Longer Query | 1391 | 1458 |
| CL-bench Life | — | 13.5% |
Writing & Preference DeepSeek V4 Pro leads
Amazon Nova Experimental Chat 12 10: 59.5 (#110), DeepSeek V4 Pro: 65.5 (#46)
| Benchmark | Amazon Nova Experimental Chat 12 10 | DeepSeek V4 Pro |
|---|---|---|
| LMArena Text | 1420 | 1451 |
| LMArena Creative Writing | 1350 | 1446 |
| LMArena Multi-Turn | 1409 | 1467 |
| EQ-Bench Creative Writing | — | 1553 |
| EQ-Bench 4 | — | 1166 |
Frequently asked questions
Is Amazon Nova Experimental Chat 12 10 better than DeepSeek V4 Pro?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 42.9 on the Noometry Index.
Is Amazon Nova Experimental Chat 12 10 or DeepSeek V4 Pro better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 42.2 in the Noometry coding category.
How many benchmarks do Amazon Nova Experimental Chat 12 10 and DeepSeek V4 Pro share?
12 benchmarks have published results for both models. Amazon Nova Experimental Chat 12 10 has 12 scored results on Noometry and DeepSeek V4 Pro has 48.