Model comparison
DeepSeek-V3.1-Terminus vs GPT-6 Astra
GPT-6 Astra is the stronger model overall, scoring 70.8 to 43.1 on the Noometry Index. DeepSeek-V3.1-Terminus costs 44× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
DeepSeekSummary
- They share 15 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and GPT-6 Astra in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 26.4.
- The biggest single-benchmark swing is LMCA: 28.6% for DeepSeek-V3.1-Terminus and 64.4% for GPT-6 Astra.
- DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
- GPT-6 Astra accepts more context: 1.05M tokens versus 164K.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1-Terminus | GPT-6 Astra | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 43.1 | 70.8 |
| Released | 2025-09-22 | 2026-09-03 |
| Weights | Open | Proprietary |
| Context window | 164K | 1.05M |
| Max output | 147K | 128K |
| Input $ / M tokens | $0.27 | $10 |
| Output $ / M tokens | $1 | $50 |
| Results tracked | 16 | 56 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Astra leads
DeepSeek-V3.1-Terminus: 42.0 (#113), GPT-6 Astra: 73.7 (#2)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| SciCode | 40.6% | 56.5% |
| LMArena Coding | 1426 | 1487 |
| ALE-Bench | 745.17 | 2,951 |
| DeepSWE | — | 74.1% |
| FrontierCode | — | 53.3% |
| LMArena WebDev | — | 1786 |
| FrontierSWE | — | 65.5% |
| GSO | — | 79.4% |
| WeirdML | — | 93.6% |
| MirrorCode | — | 46.7% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, GPT-6 Astra: 52.9 (#3)
| Benchmark | DeepSeek-V3.1-Terminus | 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
DeepSeek-V3.1-Terminus: 26.4 (#133), GPT-6 Astra: 85.1 (#1)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| CritPt | 1.7% | 31.7% |
| LMArena Hard Prompts | 1426 | 1462 |
| DTBench | 81.3% | 97.3% |
| LMCA | 28.6% | 64.4% |
| ARC-AGI-2 | — | 95% |
| Kagi LLM Benchmark | 57.4% | — |
| NYT Connections (extended) | — | 98.1% |
| ARC-AGI-1 | — | 98.5% |
| Chess Puzzles | — | 72% |
| EBR-Bench | — | 76.2% |
| Mystery Game Puzzles | — | 84% |
| Bench to the Future 3 | — | 0.14 |
| Epoch Capabilities Index | — | 166.45 |
Math GPT-6 Astra leads
DeepSeek-V3.1-Terminus: 38.5 (#137), GPT-6 Astra: 93.5 (#2)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| LMArena Math | 1402 | 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 Not comparable
DeepSeek-V3.1-Terminus: —, GPT-6 Astra: 75.3 (#1)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| GPQA Diamond | — | 95.8% |
| Humanity's Last Exam | — | 54.8% |
| SimpleQA Verified | — | 75.6% |
| Vectara Hallucination Rate | — | 8.7% |
| LMArena Expert | — | 1483 |
Multimodal Not comparable
DeepSeek-V3.1-Terminus: —, GPT-6 Astra: 55.0 (#3)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| LMArena Vision | — | 1281 |
| Blueprint-Bench 2 | — | 49.7% |
| Furniture Assembly | — | 80% |
| LMArena Document | — | 1468 |
Multilingual GPT-6 Astra leads
DeepSeek-V3.1-Terminus: 52.1 (#92), GPT-6 Astra: 53.7 (#61)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| LMArena Non-English | 1407 | 1430 |
| LMArena Russian | 1436 | 1436 |
| LMArena Chinese | — | 1484 |
| LMArena French | — | 1456 |
| LMArena German | — | 1440 |
| LMArena Japanese | — | 1379 |
| LMArena Korean | — | 1426 |
| LMArena Spanish | — | 1407 |
Instruction Following GPT-6 Astra leads
DeepSeek-V3.1-Terminus: 74.0 (#106), GPT-6 Astra: 76.3 (#44)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| LMArena Instruction Following | 1404 | 1450 |
Long Context GPT-6 Astra leads
DeepSeek-V3.1-Terminus: 43.4 (#97), GPT-6 Astra: 44.5 (#62)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| LMArena Longer Query | 1421 | 1456 |
Writing & Preference GPT-6 Astra leads
DeepSeek-V3.1-Terminus: 61.0 (#92), GPT-6 Astra: 75.3 (#7)
| Benchmark | DeepSeek-V3.1-Terminus | GPT-6 Astra |
|---|---|---|
| LMArena Text | 1419 | 1441 |
| LMArena Creative Writing | 1403 | 1418 |
| LMArena Multi-Turn | 1411 | 1448 |
| EQ-Bench Creative Writing | — | 2173 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than GPT-6 Astra?
GPT-6 Astra is the stronger model overall, scoring 70.8 to 43.1 on the Noometry Index. DeepSeek-V3.1-Terminus costs 44× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.1-Terminus or GPT-6 Astra?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; GPT-6 Astra lists at $10 and $50.
Is DeepSeek-V3.1-Terminus or GPT-6 Astra better for coding?
GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 42.0 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Astra does, with 1.05M tokens against 164K.
How many benchmarks do DeepSeek-V3.1-Terminus and GPT-6 Astra share?
15 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and GPT-6 Astra has 56.