Model comparison
DeepSeek V4 Pro vs GPT-6 Luna
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 53.3 on the Noometry Index. GPT-6 Luna costs 4.9× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Last verified . 37 shared benchmarks.
DeepSeekSummary
- They share 37 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 6 categories and GPT-6 Luna in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 64.8.
- The biggest single-benchmark swing is Mystery Game Puzzles: 43% for DeepSeek V4 Pro and 7% for GPT-6 Luna.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.66 / $1.98 for DeepSeek V4 Pro.
- GPT-6 Luna accepts more context: 1.05M tokens versus 1M.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4 Pro | GPT-6 Luna | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 54.3 | 53.3 |
| Released | 2026-04-24 | 2026-09-22 |
| Weights | Open | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 393K | 128K |
| Input $ / M tokens | $0.66 | $0.10 |
| Output $ / M tokens | $1.98 | $0.50 |
| Results tracked | 48 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
DeepSeek V4 Pro: 52.4 (#34), GPT-6 Luna: 55.5 (#25)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| FrontierCode | 28.6% | 42.4% |
| LMArena WebDev | 1582 | 1581 |
| SciCode | 51% | 54.6% |
| LMArena Coding | 1470 | 1439 |
| ALE-Bench | 1,403 | 1,577 |
| SWE-bench Verified | 77.6% | — |
| DeepSWE | — | 66.6% |
| WeirdML | 66.2% | — |
Agentic & Tool Use Too close to call
DeepSeek V4 Pro: 32.8 (#58), GPT-6 Luna: 33.3 (#54)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| APEX-Agents | 47.3% | 44.3% |
| GDP.pdf | — | 23% |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), GPT-6 Luna: 48.2 (#41)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| ARC-AGI-2 | 61.3% | 59.3% |
| NYT Connections (extended) | 91.3% | 68.7% |
| ARC-AGI-1 | 90.5% | 86.7% |
| CritPt | 18% | 19.4% |
| Chess Puzzles | 47% | 31% |
| LMArena Hard Prompts | 1461 | 1411 |
| Mystery Game Puzzles | 43% | 7% |
| DTBench | 93.9% | 90.1% |
| LMCA | 45.5% | 44.5% |
| Epoch Capabilities Index | 155.31 | 156.28 |
| Kagi LLM Benchmark | 53.5% | — |
| Surface Evolver Bench | 40% | — |
| ForecastBench | 56.1 | — |
Math GPT-6 Luna leads
DeepSeek V4 Pro: 64.8 (#30), GPT-6 Luna: 76.1 (#15)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| FrontierMath (Tiers 1-3) | 64.6% | 78.9% |
| FrontierMath Tier 4 | 26.8% | 56.1% |
| OTIS Mock AIME 2024-2025 | 98.6% | 98.9% |
| ProofBench | 50% | 64% |
| LMArena Math | 1455 | 1416 |
| MathArena Final-Answer Competitions | 76.6% | — |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), GPT-6 Luna: 57.0 (#41)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| GPQA Diamond | 91.7% | 90.5% |
| SimpleQA Verified | 52.9% | 41.4% |
| LMArena Expert | 1464 | 1444 |
| Vectara Hallucination Rate | 8.6% | — |
Multimodal Not comparable
DeepSeek V4 Pro: —, GPT-6 Luna: 42.4 (#30)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| LMArena Vision | — | 1217 |
| Blueprint-Bench 2 | — | 31.2% |
| Furniture Assembly | — | 44.2% |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), GPT-6 Luna: 50.5 (#117)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | 1439 | 1386 |
| LMArena Chinese | 1486 | 1433 |
| LMArena French | 1472 | 1420 |
| LMArena German | 1458 | 1369 |
| LMArena Japanese | 1445 | 1369 |
| LMArena Korean | 1447 | 1360 |
| LMArena Russian | 1453 | 1394 |
| LMArena Spanish | 1458 | 1393 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), GPT-6 Luna: 74.3 (#99)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | 1448 | 1409 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), GPT-6 Luna: 43.0 (#111)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | 1458 | 1409 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), GPT-6 Luna: 58.3 (#119)
| Benchmark | DeepSeek V4 Pro | GPT-6 Luna |
|---|---|---|
| LMArena Text | 1451 | 1391 |
| LMArena Creative Writing | 1446 | 1363 |
| LMArena Multi-Turn | 1467 | 1396 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than GPT-6 Luna?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 53.3 on the Noometry Index. GPT-6 Luna costs 4.9× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4 Pro or GPT-6 Luna?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; DeepSeek V4 Pro lists at $0.66 and $1.98.
Is DeepSeek V4 Pro or GPT-6 Luna better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 52.4 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 1M.
How many benchmarks do DeepSeek V4 Pro and GPT-6 Luna share?
37 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and GPT-6 Luna has 42.