Model comparison
DeepSeek V4.1 Flash vs GPT-5
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 50.9 on the Noometry Index.
Last verified . 32 shared benchmarks.
DeepSeekSummary
- They share 32 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 7 categories and GPT-5 in 3 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in long context, where GPT-5 leads 69.5 to 45.2.
- The biggest single-benchmark swing is ProofBench: 54% for DeepSeek V4.1 Flash and 18% for GPT-5.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.25 / $10 for GPT-5.
- 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 | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 52.8 | 50.9 |
| Released | 2026-09-09 | 2025-08-07 |
| 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 | 69 |
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: 50.3 (#47)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| LMArena WebDev | 1619 | 1418 |
| SciCode | 51.9% | 42.9% |
| LMArena Coding | 1506 | 1436 |
| ALE-Bench | 1,092 | 1,162 |
| SWE-bench Verified | — | 73.6% |
| SWE-bench Verified (bash only) | — | 65% |
| Aider Polyglot | — | 88% |
| GSO | — | 6.9% |
| WeirdML | — | 60.7% |
| AlgoTune | — | 1.67 |
Agentic & Tool Use GPT-5 leads
DeepSeek V4.1 Flash: 31.2 (#69), GPT-5: 33.1 (#56)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| Terminal-Bench | — | 49.6% |
| APEX-Agents | 39.5% | — |
| GDPval | — | 34.8% |
| Remote Labor Index | — | 1.7% |
| DeepResearch Bench | — | 49.6% |
| BALROG | — | 32.8% |
| GDP.pdf | 19.8% | — |
| LMArena Search | — | 1133 |
| METR Time Horizons | — | 69.6% |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), GPT-5: 38.3 (#64)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| CritPt | 14.3% | 12.6% |
| LMArena Hard Prompts | 1483 | 1416 |
| Mystery Game Puzzles | 43% | 23% |
| DTBench | 89.9% | 90.7% |
| LMCA | 47% | 40% |
| Epoch Capabilities Index | 154.9 | 150 |
| ARC-AGI-2 | — | 9.9% |
| SimpleBench | — | 56.7% |
| Kagi LLM Benchmark | — | 72.7% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 65.7% |
| Chess Puzzles | — | 37% |
| EnigmaEval | — | 10.5% |
| EBR-Bench | — | 12.7% |
| Surface Evolver Bench | 46.3% | — |
| ForecastBench | — | 61.4 |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), GPT-5: 55.0 (#44)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 55.4% |
| FrontierMath Tier 4 | 26.8% | 22% |
| OTIS Mock AIME 2024-2025 | 98.3% | 91.4% |
| ProofBench | 54% | 18% |
| LMArena Math | 1477 | 1407 |
| Omni-MATH | — | 64.7% |
| MATH Level 5 | — | 98.1% |
| FrontierMath (Feb 2025 set) | — | 32.4% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), GPT-5: 56.6 (#43)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| GPQA Diamond | 89.8% | 86.2% |
| LMArena Expert | 1506 | 1419 |
| Humanity's Last Exam | — | 25.3% |
| SimpleQA Verified | — | 50.1% |
| MMLU-Pro | — | 86.3% |
| Confabulations | — | 10.3% |
| Vectara Hallucination Rate | — | 14.7% |
| GPQA (HELM) | — | 79.2% |
Multimodal GPT-5 leads
DeepSeek V4.1 Flash: 39.1 (#61), GPT-5: 46.8 (#13)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| LMArena Vision | 1277 | 1232 |
| GeoBench | — | 81% |
| VPCT | — | 66% |
| Furniture Assembly | 34.2% | — |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), GPT-5: 51.4 (#110)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| LMArena Non-English | 1448 | 1397 |
| LMArena Chinese | 1497 | 1422 |
| LMArena French | 1452 | 1410 |
| LMArena German | 1484 | 1416 |
| LMArena Japanese | 1412 | 1409 |
| LMArena Korean | 1452 | 1360 |
| LMArena Russian | 1471 | 1406 |
| LMArena Spanish | 1459 | 1399 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.3 (#26), GPT-5: 73.8 (#113)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| LMArena Instruction Following | 1474 | 1388 |
| IFEval | — | 87.5% |
Long Context GPT-5 leads
DeepSeek V4.1 Flash: 45.2 (#47), GPT-5: 69.5 (#2)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| LMArena Longer Query | 1475 | 1399 |
| Fiction.LiveBench | — | 97.2% |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 65.4 (#48), GPT-5: 63.4 (#65)
| Benchmark | DeepSeek V4.1 Flash | GPT-5 |
|---|---|---|
| LMArena Text | 1462 | 1406 |
| LMArena Creative Writing | 1435 | 1365 |
| EQ-Bench Creative Writing | 1540 | 1627 |
| LMArena Multi-Turn | 1457 | 1426 |
| Short-Story Creative Writing | — | 86% |
| WildBench | — | 85.7% |
Frequently asked questions
Is DeepSeek V4.1 Flash better than GPT-5?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 50.9 on the Noometry Index.
Which is cheaper, DeepSeek V4.1 Flash or GPT-5?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5 lists at $1.25 and $10.
Is DeepSeek V4.1 Flash or GPT-5 better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 50.3 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 share?
32 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and GPT-5 has 69.