Model comparison
GPT-6 Astra vs Solar Pro4
GPT-6 Astra is the stronger model overall, scoring 70.8 to 42.1 on the Noometry Index. Solar Pro4 costs 38× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
UpstageSummary
- They share 18 benchmarks with published results for both. GPT-6 Astra scores higher in 8 categories and Solar Pro4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 28.5.
- Solar Pro4 is cheaper at $0.30 / $1.20 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
- GPT-6 Astra accepts more context: 1.05M tokens versus 524K.
Side by side
| GPT-6 Astra | Solar Pro4 | |
|---|---|---|
| Provider | OpenAI | Upstage |
| Noometry Index | 70.8 | 42.1 |
| Released | 2026-09-03 | 2026-08-06 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 524K |
| Max output | 128K | 131K |
| Input $ / M tokens | $10 | $0.30 |
| Output $ / M tokens | $50 | $1.20 |
| Results tracked | 56 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Astra leads
GPT-6 Astra: 73.7 (#2), Solar Pro4: 40.1 (#149)
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena WebDev | 1786 | 1371 |
| LMArena Coding | 1487 | 1437 |
| DeepSWE | 74.1% | — |
| FrontierCode | 53.3% | — |
| FrontierSWE | 65.5% | — |
| SciCode | 56.5% | — |
| GSO | 79.4% | — |
| WeirdML | 93.6% | — |
| MirrorCode | 46.7% | — |
| ALE-Bench | 2,951 | — |
Agentic & Tool Use Not comparable
GPT-6 Astra: 52.9 (#3), Solar Pro4: —
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| 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
GPT-6 Astra: 85.1 (#1), Solar Pro4: 28.5 (#104)
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena Hard Prompts | 1462 | 1399 |
| 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
GPT-6 Astra: 93.5 (#2), Solar Pro4: 38.8 (#128)
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena Math | 1465 | 1416 |
| 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
GPT-6 Astra: 75.3 (#1), Solar Pro4: 39.8 (#129)
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena Expert | 1483 | 1427 |
| GPQA Diamond | 95.8% | — |
| Humanity's Last Exam | 54.8% | — |
| SimpleQA Verified | 75.6% | — |
| Vectara Hallucination Rate | 8.7% | — |
Multimodal Not comparable
GPT-6 Astra: 55.0 (#3), Solar Pro4: —
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena Vision | 1281 | — |
| Blueprint-Bench 2 | 49.7% | — |
| Furniture Assembly | 80% | — |
| LMArena Document | 1468 | — |
Multilingual GPT-6 Astra leads
GPT-6 Astra: 53.7 (#61), Solar Pro4: 48.7 (#139)
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena Non-English | 1430 | 1361 |
| LMArena Chinese | 1484 | 1415 |
| LMArena French | 1456 | 1397 |
| LMArena German | 1440 | 1364 |
| LMArena Japanese | 1379 | 1309 |
| LMArena Korean | 1426 | 1382 |
| LMArena Russian | 1436 | 1360 |
| LMArena Spanish | 1407 | 1401 |
Instruction Following GPT-6 Astra leads
GPT-6 Astra: 76.3 (#44), Solar Pro4: 72.7 (#132)
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena Instruction Following | 1450 | 1377 |
Long Context GPT-6 Astra leads
GPT-6 Astra: 44.5 (#62), Solar Pro4: 42.1 (#130)
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena Longer Query | 1456 | 1381 |
Writing & Preference GPT-6 Astra leads
GPT-6 Astra: 75.3 (#7), Solar Pro4: 56.5 (#138)
| Benchmark | GPT-6 Astra | Solar Pro4 |
|---|---|---|
| LMArena Text | 1441 | 1386 |
| LMArena Creative Writing | 1418 | 1316 |
| LMArena Multi-Turn | 1448 | 1385 |
| EQ-Bench Creative Writing | 2173 | — |
Frequently asked questions
Is GPT-6 Astra better than Solar Pro4?
GPT-6 Astra is the stronger model overall, scoring 70.8 to 42.1 on the Noometry Index. Solar Pro4 costs 38× 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 Astra or Solar Pro4?
Solar Pro4 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-6 Astra lists at $10 and $50.
Is GPT-6 Astra or Solar Pro4 better for coding?
GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Astra does, with 1.05M tokens against 524K.
How many benchmarks do GPT-6 Astra and Solar Pro4 share?
18 benchmarks have published results for both models. GPT-6 Astra has 56 scored results on Noometry and Solar Pro4 has 18.