Model comparison
DeepSeek V4.1 Flash vs GPT-5.1
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 49.0 on the Noometry Index.
Last verified . 28 shared benchmarks.
DeepSeekSummary
- They share 28 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 6 categories and GPT-5.1 in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 52.2.
- The biggest single-benchmark swing is Mystery Game Puzzles: 43% for DeepSeek V4.1 Flash and 19% for GPT-5.1.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 400K.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4.1 Flash | GPT-5.1 | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 52.8 | 49.0 |
| Released | 2026-09-09 | 2025-11-13 |
| Weights | Open | Proprietary |
| Context window | 1M | 400K |
| Max output | 393K | 128K |
| Input $ / M tokens | $0.15 | $1.25 |
| Output $ / M tokens | $0.60 | $10 |
| Results tracked | 37 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 52.9 (#32), GPT-5.1: 46.4 (#66)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| LMArena WebDev | 1619 | 1395 |
| SciCode | 51.9% | 43.3% |
| LMArena Coding | 1506 | 1454 |
| ALE-Bench | 1,092 | 1,192 |
| SWE-bench Verified | — | 68% |
| SWE-bench Verified (bash only) | — | 66% |
| GSO | — | 13.7% |
| WeirdML | — | 60.8% |
| LiveBench Coding | — | 72.5% |
Agentic & Tool Use GPT-5.1 leads
DeepSeek V4.1 Flash: 31.2 (#69), GPT-5.1: 32.7 (#60)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| Terminal-Bench | — | 47.6% |
| APEX-Agents | 39.5% | — |
| DeepResearch Bench | — | 42.8% |
| GDP.pdf | 19.8% | — |
| LMArena Search | — | 1199 |
| Vending-Bench 2 | — | 1,473 |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), GPT-5.1: 39.8 (#58)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| CritPt | 14.3% | 4.9% |
| LMArena Hard Prompts | 1483 | 1457 |
| Mystery Game Puzzles | 43% | 19% |
| DTBench | 89.9% | 90.1% |
| LMCA | 47% | 43.9% |
| Epoch Capabilities Index | 154.9 | 149.64 |
| ARC-AGI-2 | — | 17.6% |
| SimpleBench | — | 53.2% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 72.8% |
| Chess Puzzles | — | 32% |
| EnigmaEval | — | 11.2% |
| LiveBench Reasoning | — | 95.8% |
| LiveBench Data Analysis | — | 72.1% |
| Surface Evolver Bench | 46.3% | — |
| ForecastBench | — | 58.1 |
| LiveBench | — | 78.8% |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), GPT-5.1: 52.2 (#51)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 88.6% |
| LMArena Math | 1477 | 1447 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 26.8% | — |
| ProofBench | 54% | — |
| Omni-MATH | — | 46.4% |
| LiveBench Math | — | 94.5% |
| FrontierMath (Feb 2025 set) | — | 31% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), GPT-5.1: 50.6 (#71)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 89.8% | 87.6% |
| LMArena Expert | 1506 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| SimpleQA Verified | — | 48% |
| MMLU-Pro | — | 57.9% |
| Vectara Hallucination Rate | — | 10.9% |
| GPQA (HELM) | — | 44.2% |
Multimodal GPT-5.1 leads
DeepSeek V4.1 Flash: 39.1 (#61), GPT-5.1: 44.8 (#19)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1277 | 1250 |
| VPCT | — | 58.7% |
| Furniture Assembly | 34.2% | — |
| LMArena Document | — | 1403 |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), GPT-5.1: 53.8 (#56)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1448 | 1431 |
| LMArena Chinese | 1497 | 1495 |
| LMArena French | 1452 | 1450 |
| LMArena German | 1484 | 1438 |
| LMArena Japanese | 1412 | 1453 |
| LMArena Korean | 1452 | 1401 |
| LMArena Russian | 1471 | 1435 |
| LMArena Spanish | 1459 | 1433 |
Instruction Following GPT-5.1 leads
DeepSeek V4.1 Flash: 77.3 (#26), GPT-5.1: 83.9 (#1)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| LMArena Instruction Following | 1474 | 1443 |
| LiveBench Instruction Following | — | 93.3% |
| IFEval | — | 93.5% |
Long Context GPT-5.1 leads
DeepSeek V4.1 Flash: 45.2 (#47), GPT-5.1: 47.6 (#14)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1475 | 1447 |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference Too close to call
DeepSeek V4.1 Flash: 65.4 (#48), GPT-5.1: 64.5 (#55)
| Benchmark | DeepSeek V4.1 Flash | GPT-5.1 |
|---|---|---|
| LMArena Text | 1462 | 1443 |
| LMArena Creative Writing | 1435 | 1427 |
| LMArena Multi-Turn | 1457 | 1450 |
| EQ-Bench Creative Writing | 1540 | — |
| WildBench | — | 86.3% |
| LiveBench Language | — | 80.2% |
Frequently asked questions
Is DeepSeek V4.1 Flash better than GPT-5.1?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 49.0 on the Noometry Index.
Which is cheaper, DeepSeek V4.1 Flash or GPT-5.1?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5.1 lists at $1.25 and $10.
Is DeepSeek V4.1 Flash or GPT-5.1 better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 46.4 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4.1 Flash does, with 1M tokens against 400K.
How many benchmarks do DeepSeek V4.1 Flash and GPT-5.1 share?
28 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and GPT-5.1 has 63.