Model comparison
DeepSeek V4.1 Flash vs GPT-6 Luna
DeepSeek V4.1 Flash and GPT-6 Luna score almost the same on the Noometry Index (52.8 vs 53.3), so choose on price, context window or the category you care about most.
Last verified . 35 shared benchmarks.
DeepSeekSummary
- They share 35 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 6 categories and GPT-6 Luna in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 66.7.
- The biggest single-benchmark swing is Mystery Game Puzzles: 43% for DeepSeek V4.1 Flash and 7% for GPT-6 Luna.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.15 / $0.60 for DeepSeek V4.1 Flash.
- GPT-6 Luna accepts more context: 1.05M tokens versus 1M.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4.1 Flash | GPT-6 Luna | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 52.8 | 53.3 |
| Released | 2026-09-09 | 2026-09-22 |
| Weights | Open | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 393K | 128K |
| Input $ / M tokens | $0.15 | $0.10 |
| Output $ / M tokens | $0.60 | $0.50 |
| Results tracked | 37 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
DeepSeek V4.1 Flash: 52.9 (#32), GPT-6 Luna: 55.5 (#25)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| LMArena WebDev | 1619 | 1581 |
| SciCode | 51.9% | 54.6% |
| LMArena Coding | 1506 | 1439 |
| ALE-Bench | 1,092 | 1,577 |
| DeepSWE | — | 66.6% |
| FrontierCode | — | 42.4% |
Agentic & Tool Use GPT-6 Luna leads
DeepSeek V4.1 Flash: 31.2 (#69), GPT-6 Luna: 33.3 (#54)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| APEX-Agents | 39.5% | 44.3% |
| GDP.pdf | 19.8% | 23% |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), GPT-6 Luna: 48.2 (#41)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| NYT Connections (extended) | 89.6% | 68.7% |
| CritPt | 14.3% | 19.4% |
| LMArena Hard Prompts | 1483 | 1411 |
| Mystery Game Puzzles | 43% | 7% |
| DTBench | 89.9% | 90.1% |
| LMCA | 47% | 44.5% |
| Epoch Capabilities Index | 154.9 | 156.28 |
| ARC-AGI-2 | — | 59.3% |
| ARC-AGI-1 | — | 86.7% |
| Chess Puzzles | — | 31% |
| Surface Evolver Bench | 46.3% | — |
Math GPT-6 Luna leads
DeepSeek V4.1 Flash: 66.7 (#25), GPT-6 Luna: 76.1 (#15)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 78.9% |
| FrontierMath Tier 4 | 26.8% | 56.1% |
| OTIS Mock AIME 2024-2025 | 98.3% | 98.9% |
| ProofBench | 54% | 64% |
| LMArena Math | 1477 | 1416 |
Knowledge Too close to call
DeepSeek V4.1 Flash: 57.9 (#38), GPT-6 Luna: 57.0 (#41)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| GPQA Diamond | 89.8% | 90.5% |
| LMArena Expert | 1506 | 1444 |
| SimpleQA Verified | — | 41.4% |
Multimodal GPT-6 Luna leads
DeepSeek V4.1 Flash: 39.1 (#61), GPT-6 Luna: 42.4 (#30)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| LMArena Vision | 1277 | 1217 |
| Furniture Assembly | 34.2% | 44.2% |
| Blueprint-Bench 2 | — | 31.2% |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), GPT-6 Luna: 50.5 (#117)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | 1448 | 1386 |
| LMArena Chinese | 1497 | 1433 |
| LMArena French | 1452 | 1420 |
| LMArena German | 1484 | 1369 |
| LMArena Japanese | 1412 | 1369 |
| LMArena Korean | 1452 | 1360 |
| LMArena Russian | 1471 | 1394 |
| LMArena Spanish | 1459 | 1393 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.3 (#26), GPT-6 Luna: 74.3 (#99)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | 1474 | 1409 |
Long Context DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 45.2 (#47), GPT-6 Luna: 43.0 (#111)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | 1475 | 1409 |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 65.4 (#48), GPT-6 Luna: 58.3 (#119)
| Benchmark | DeepSeek V4.1 Flash | GPT-6 Luna |
|---|---|---|
| LMArena Text | 1462 | 1391 |
| LMArena Creative Writing | 1435 | 1363 |
| LMArena Multi-Turn | 1457 | 1396 |
| EQ-Bench Creative Writing | 1540 | — |
Frequently asked questions
Is DeepSeek V4.1 Flash better than GPT-6 Luna?
DeepSeek V4.1 Flash and GPT-6 Luna score almost the same on the Noometry Index (52.8 vs 53.3), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek V4.1 Flash 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 V4.1 Flash lists at $0.15 and $0.60.
Is DeepSeek V4.1 Flash or GPT-6 Luna better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 52.9 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 V4.1 Flash and GPT-6 Luna share?
35 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and GPT-6 Luna has 42.