Model comparison
DeepSeek-R1 vs GPT-5.4 nano
DeepSeek-R1 and GPT-5.4 nano score almost the same on the Noometry Index (42.0 vs 41.9), so choose on price, context window or the category you care about most.
Last verified . 29 shared benchmarks.
DeepSeekSummary
- They share 29 benchmarks with published results for both. DeepSeek-R1 scores higher in 7 categories and GPT-5.4 nano in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-R1 leads 61.5 to 55.9.
- The biggest single-benchmark swing is ARC-AGI-1: 21.2% for DeepSeek-R1 and 51.5% for GPT-5.4 nano.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $0.50 / $2.15 for DeepSeek-R1.
- GPT-5.4 nano accepts more context: 400K tokens versus 164K.
Side by side
| DeepSeek-R1 | GPT-5.4 nano | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 42.0 | 41.9 |
| Released | 2025-01-20 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | 164K | 400K |
| Max output | 64K | 128K |
| Input $ / M tokens | $0.50 | $0.20 |
| Output $ / M tokens | $2.15 | $1.25 |
| Results tracked | 52 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 45.9 (#63), GPT-5.4 nano: 43.1 (#82)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| SciCode | 35.7% | 46.9% |
| WeirdML | 41.6% | 49.2% |
| LMArena Coding | 1427 | 1404 |
| ALE-Bench | 804.12 | 1,005 |
| Aider Polyglot | 71.4% | — |
| LiveBench Coding | 66.7% | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use Not comparable
DeepSeek-R1: 30.4 (#68), GPT-5.4 nano: —
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
Reasoning GPT-5.4 nano leads
DeepSeek-R1: 18.7 (#253), GPT-5.4 nano: 23.8 (#159)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| ARC-AGI-2 | 1.3% | 5.7% |
| Kagi LLM Benchmark | 69.4% | 39.7% |
| ARC-AGI-1 | 21.2% | 51.5% |
| CritPt | 1.1% | 9.3% |
| LMArena Hard Prompts | 1415 | 1380 |
| Epoch Capabilities Index | 141.3 | 145.78 |
| ForecastBench | 60 | 57.3 |
| SimpleBench | 40.8% | — |
| Chess Puzzles | — | 30% |
| LiveBench Reasoning | 83.2% | — |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 36.9% |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 42.0 (#70), GPT-5.4 nano: 40.7 (#76)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 87.8% |
| LMArena Math | 1401 | 1406 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 5% |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |
| FrontierMath (Feb 2025 set) | — | 25.9% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge DeepSeek-R1 leads
DeepSeek-R1: 44.3 (#81), GPT-5.4 nano: 41.8 (#94)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| GPQA Diamond | 76.3% | 78.5% |
| Vectara Hallucination Rate | 11.3% | 3.1% |
| LMArena Expert | 1394 | 1397 |
| SimpleQA Verified | — | 11.7% |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| GPQA (HELM) | 66.6% | — |
Multimodal Not comparable
DeepSeek-R1: —, GPT-5.4 nano: 36.0 (#73)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | — | 1196 |
Multilingual DeepSeek-R1 leads
DeepSeek-R1: 52.6 (#80), GPT-5.4 nano: 48.8 (#136)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1412 | 1359 |
| LMArena Chinese | 1443 | 1391 |
| LMArena French | 1415 | 1396 |
| LMArena German | 1405 | 1368 |
| LMArena Japanese | 1394 | 1342 |
| LMArena Korean | 1359 | 1321 |
| LMArena Russian | 1423 | 1363 |
| LMArena Spanish | 1410 | 1368 |
Instruction Following Too close to call
DeepSeek-R1: 72.2 (#138), GPT-5.4 nano: 72.1 (#139)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1382 | 1362 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), GPT-5.4 nano: 41.6 (#131)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1391 | 1365 |
| Fiction.LiveBench | 75% | — |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.5 (#84), GPT-5.4 nano: 55.9 (#137)
| Benchmark | DeepSeek-R1 | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1428 | 1372 |
| LMArena Creative Writing | 1406 | 1313 |
| LMArena Multi-Turn | 1404 | 1382 |
| Short-Story Creative Writing | 83% | — |
| EQ-Bench Creative Writing | 1500 | — |
| WildBench | 82.8% | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than GPT-5.4 nano?
DeepSeek-R1 and GPT-5.4 nano score almost the same on the Noometry Index (42.0 vs 41.9), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-R1 or GPT-5.4 nano?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; DeepSeek-R1 lists at $0.50 and $2.15.
Is DeepSeek-R1 or GPT-5.4 nano better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 45.9 versus 43.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 164K.
How many benchmarks do DeepSeek-R1 and GPT-5.4 nano share?
29 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and GPT-5.4 nano has 40.