Model comparison
Amazon Nova Experimental Chat 10 20 vs GPT-6 Astra
GPT-6 Astra is the stronger model overall, scoring 70.8 to 42.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
Summary
- They share 17 benchmarks with published results for both. Amazon Nova Experimental Chat 10 20 scores higher in 0 categories and GPT-6 Astra in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 28.4.
Side by side
| Amazon Nova Experimental Chat 10 20 | GPT-6 Astra | |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 42.1 | 70.8 |
| Released | — | 2026-09-03 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 128K |
| Input $ / M tokens | — | $10 |
| Output $ / M tokens | — | $50 |
| Results tracked | 17 | 56 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Astra leads
Amazon Nova Experimental Chat 10 20: 41.6 (#123), GPT-6 Astra: 73.7 (#2)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Coding | 1412 | 1487 |
| DeepSWE | — | 74.1% |
| FrontierCode | — | 53.3% |
| LMArena WebDev | — | 1786 |
| FrontierSWE | — | 65.5% |
| SciCode | — | 56.5% |
| GSO | — | 79.4% |
| WeirdML | — | 93.6% |
| MirrorCode | — | 46.7% |
| ALE-Bench | — | 2,951 |
Agentic & Tool Use Not comparable
Amazon Nova Experimental Chat 10 20: —, GPT-6 Astra: 52.9 (#3)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| APEX-Agents | — | 64.7% |
| Remote Labor Index | — | 20.8% |
| BALROG | — | 68.3% |
| GDP.pdf | — | 34.2% |
| Vending-Bench 2 | — | 15,515 |
Reasoning GPT-6 Astra leads
Amazon Nova Experimental Chat 10 20: 28.4 (#106), GPT-6 Astra: 85.1 (#1)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Hard Prompts | 1396 | 1462 |
| ARC-AGI-2 | — | 95% |
| NYT Connections (extended) | — | 98.1% |
| ARC-AGI-1 | — | 98.5% |
| CritPt | — | 31.7% |
| Chess Puzzles | — | 72% |
| EBR-Bench | — | 76.2% |
| Mystery Game Puzzles | — | 84% |
| DTBench | — | 97.3% |
| LMCA | — | 64.4% |
| Bench to the Future 3 | — | 0.14 |
| Epoch Capabilities Index | — | 166.45 |
Math GPT-6 Astra leads
Amazon Nova Experimental Chat 10 20: 39.0 (#119), GPT-6 Astra: 93.5 (#2)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Math | 1425 | 1465 |
| FrontierMath (Tiers 1-3) | — | 93.7% |
| FrontierMath Tier 4 | — | 97.6% |
| OTIS Mock AIME 2024-2025 | — | 100% |
| ProofBench | — | 99% |
| FrontierMath Erdős | — | 2.9% |
Knowledge GPT-6 Astra leads
Amazon Nova Experimental Chat 10 20: 38.5 (#144), GPT-6 Astra: 75.3 (#1)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Expert | 1386 | 1483 |
| GPQA Diamond | — | 95.8% |
| Humanity's Last Exam | — | 54.8% |
| SimpleQA Verified | — | 75.6% |
| Vectara Hallucination Rate | — | 8.7% |
Multimodal Not comparable
Amazon Nova Experimental Chat 10 20: —, GPT-6 Astra: 55.0 (#3)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Vision | — | 1281 |
| Blueprint-Bench 2 | — | 49.7% |
| Furniture Assembly | — | 80% |
| LMArena Document | — | 1468 |
Multilingual GPT-6 Astra leads
Amazon Nova Experimental Chat 10 20: 49.5 (#131), GPT-6 Astra: 53.7 (#61)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Non-English | 1372 | 1430 |
| LMArena Chinese | 1404 | 1484 |
| LMArena French | 1425 | 1456 |
| LMArena German | 1391 | 1440 |
| LMArena Japanese | 1360 | 1379 |
| LMArena Korean | 1324 | 1426 |
| LMArena Russian | 1371 | 1436 |
| LMArena Spanish | 1382 | 1407 |
Instruction Following GPT-6 Astra leads
Amazon Nova Experimental Chat 10 20: 72.1 (#141), GPT-6 Astra: 76.3 (#44)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Instruction Following | 1365 | 1450 |
Long Context GPT-6 Astra leads
Amazon Nova Experimental Chat 10 20: 41.7 (#133), GPT-6 Astra: 44.5 (#62)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Longer Query | 1370 | 1456 |
Writing & Preference GPT-6 Astra leads
Amazon Nova Experimental Chat 10 20: 56.6 (#136), GPT-6 Astra: 75.3 (#7)
| Benchmark | Amazon Nova Experimental Chat 10 20 | GPT-6 Astra |
|---|---|---|
| LMArena Text | 1394 | 1441 |
| LMArena Creative Writing | 1318 | 1418 |
| LMArena Multi-Turn | 1364 | 1448 |
| EQ-Bench Creative Writing | — | 2173 |
Frequently asked questions
Is Amazon Nova Experimental Chat 10 20 better than GPT-6 Astra?
GPT-6 Astra is the stronger model overall, scoring 70.8 to 42.1 on the Noometry Index.
Is Amazon Nova Experimental Chat 10 20 or GPT-6 Astra better for coding?
GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 41.6 in the Noometry coding category.
How many benchmarks do Amazon Nova Experimental Chat 10 20 and GPT-6 Astra share?
17 benchmarks have published results for both models. Amazon Nova Experimental Chat 10 20 has 17 scored results on Noometry and GPT-6 Astra has 56.