Model comparison
GPT-6.1 Sol vs GPT-6 Astra
GPT-6 Astra is the stronger model overall, scoring 70.8 to 65.6 on the Noometry Index. GPT-6.1 Sol costs 5.0× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. GPT-6.1 Sol scores higher in 4 categories and GPT-6 Astra in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-6 Astra leads 52.9 to 39.6.
- The biggest single-benchmark swing is EBR-Bench: 54.3% for GPT-6.1 Sol and 76.2% for GPT-6 Astra.
- GPT-6.1 Sol is cheaper at $2 / $10 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
Side by side
| GPT-6.1 Sol | GPT-6 Astra | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 65.6 | 70.8 |
| Released | 2026-09-29 | 2026-09-03 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $2 | $10 |
| Output $ / M tokens | $10 | $50 |
| Results tracked | 34 | 56 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Astra leads
GPT-6.1 Sol: 63.2 (#8), GPT-6 Astra: 73.7 (#2)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| DeepSWE | 75.2% | 74.1% |
| FrontierCode | 50.2% | 53.3% |
| LMArena WebDev | 1755 | 1786 |
| SciCode | 55.8% | 56.5% |
| LMArena Coding | 1487 | 1487 |
| FrontierSWE | — | 65.5% |
| GSO | — | 79.4% |
| WeirdML | — | 93.6% |
| MirrorCode | — | 46.7% |
| ALE-Bench | — | 2,951 |
Agentic & Tool Use GPT-6 Astra leads
GPT-6.1 Sol: 39.6 (#26), GPT-6 Astra: 52.9 (#3)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| APEX-Agents | 60% | 64.7% |
| GDP.pdf | 32% | 34.2% |
| Remote Labor Index | — | 20.8% |
| BALROG | — | 68.3% |
| Vending-Bench 2 | — | 15,515 |
Reasoning GPT-6 Astra leads
GPT-6.1 Sol: 81.9 (#2), GPT-6 Astra: 85.1 (#1)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| ARC-AGI-2 | 94.2% | 95% |
| NYT Connections (extended) | 95.5% | 98.1% |
| ARC-AGI-1 | 98.5% | 98.5% |
| CritPt | 31.7% | 31.7% |
| Chess Puzzles | 61% | 72% |
| EBR-Bench | 54.3% | 76.2% |
| LMArena Hard Prompts | 1466 | 1462 |
| Mystery Game Puzzles | 80% | 84% |
| Epoch Capabilities Index | 166.09 | 166.45 |
| DTBench | — | 97.3% |
| LMCA | — | 64.4% |
| Bench to the Future 3 | — | 0.14 |
Math Too close to call
GPT-6.1 Sol: 93.7 (#1), GPT-6 Astra: 93.5 (#2)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| FrontierMath (Tiers 1-3) | 93.7% | 93.7% |
| FrontierMath Tier 4 | 100% | 97.6% |
| OTIS Mock AIME 2024-2025 | 100% | 100% |
| ProofBench | 99% | 99% |
| LMArena Math | 1464 | 1465 |
| FrontierMath Erdős | — | 2.9% |
Knowledge GPT-6 Astra leads
GPT-6.1 Sol: 71.8 (#4), GPT-6 Astra: 75.3 (#1)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| GPQA Diamond | 95.4% | 95.8% |
| SimpleQA Verified | 73.9% | 75.6% |
| LMArena Expert | 1502 | 1483 |
| Humanity's Last Exam | — | 54.8% |
| Vectara Hallucination Rate | — | 8.7% |
Multimodal GPT-6 Astra leads
GPT-6.1 Sol: 52.7 (#5), GPT-6 Astra: 55.0 (#3)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| LMArena Vision | 1288 | 1281 |
| Furniture Assembly | 80% | 80% |
| Blueprint-Bench 2 | — | 49.7% |
| LMArena Document | — | 1468 |
Multilingual Too close to call
GPT-6.1 Sol: 54.3 (#46), GPT-6 Astra: 53.7 (#61)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| LMArena Non-English | 1438 | 1430 |
| LMArena Chinese | 1477 | 1484 |
| LMArena Russian | 1455 | 1436 |
| LMArena French | — | 1456 |
| LMArena German | — | 1440 |
| LMArena Japanese | — | 1379 |
| LMArena Korean | — | 1426 |
| LMArena Spanish | — | 1407 |
Instruction Following Too close to call
GPT-6.1 Sol: 77.0 (#29), GPT-6 Astra: 76.3 (#44)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| LMArena Instruction Following | 1468 | 1450 |
Long Context Too close to call
GPT-6.1 Sol: 44.9 (#54), GPT-6 Astra: 44.5 (#62)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| LMArena Longer Query | 1465 | 1456 |
Writing & Preference GPT-6 Astra leads
GPT-6.1 Sol: 63.6 (#63), GPT-6 Astra: 75.3 (#7)
| Benchmark | GPT-6.1 Sol | GPT-6 Astra |
|---|---|---|
| LMArena Text | 1447 | 1441 |
| LMArena Creative Writing | 1432 | 1418 |
| LMArena Multi-Turn | 1449 | 1448 |
| EQ-Bench Creative Writing | — | 2173 |
Frequently asked questions
Is GPT-6.1 Sol better than GPT-6 Astra?
GPT-6 Astra is the stronger model overall, scoring 70.8 to 65.6 on the Noometry Index. GPT-6.1 Sol costs 5.0× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Which is cheaper, GPT-6.1 Sol or GPT-6 Astra?
GPT-6.1 Sol is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; GPT-6 Astra lists at $10 and $50.
Is GPT-6.1 Sol or GPT-6 Astra better for coding?
GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 63.2 in the Noometry coding category.
Which has the bigger context window?
Both accept 1.05M tokens.
How many benchmarks do GPT-6.1 Sol and GPT-6 Astra share?
34 benchmarks have published results for both models. GPT-6.1 Sol has 34 scored results on Noometry and GPT-6 Astra has 56.