Model comparison
DeepSeek-R1 vs GPT-4 Turbo
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 30.5 on the Noometry Index.
Last verified . 26 shared benchmarks.
DeepSeekSummary
- They share 26 benchmarks with published results for both. DeepSeek-R1 scores higher in 8 categories and GPT-4 Turbo in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-R1 leads 43.8 to 9.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.4% for DeepSeek-R1 and 6.7% for GPT-4 Turbo.
- DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- DeepSeek-R1 accepts more context: 164K tokens versus 128K.
Side by side
| DeepSeek-R1 | GPT-4 Turbo | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 42.3 | 30.5 |
| Released | 2025-01-20 | 2023-11-06 |
| Weights | Proprietary | Proprietary |
| Context window | 164K | 128K |
| Max output | 64K | 4K |
| Input $ / M tokens | $0.50 | $10 |
| Output $ / M tokens | $2.15 | $30 |
| Results tracked | 52 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), GPT-4 Turbo: 33.8 (#249)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| WeirdML | 41.6% | 18% |
| LMArena Coding | 1427 | 1268 |
| Aider Polyglot | 71.4% | — |
| SciCode | 35.7% | — |
| BigCodeBench Instruct | — | 48.2% |
| LiveBench Coding | 66.7% | — |
| BigCodeBench Complete | — | 58.2% |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
| HumanEval+ | — | 86.6% |
| MBPP+ | — | 73.3% |
Agentic & Tool Use Not comparable
DeepSeek-R1: 30.7 (#75), GPT-4 Turbo: —
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| METR Time Horizons | 53.8% | 36.7% |
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
Reasoning DeepSeek-R1 leads
DeepSeek-R1: 18.6 (#278), GPT-4 Turbo: 15.3 (#317)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| SimpleBench | 40.8% | 25.1% |
| LMArena Hard Prompts | 1416 | 1251 |
| Epoch Capabilities Index | 141.29 | 127.25 |
| ForecastBench | 60 | 59.4 |
| ARC-AGI-2 | 1.3% | — |
| Kagi LLM Benchmark | 69.4% | — |
| ARC-AGI-1 | 21.2% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 6% |
| LiveBench Reasoning | 83.2% | — |
| DTBench | — | 61.6% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 9.8% |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), GPT-4 Turbo: 9.0 (#322)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 6.7% |
| LMArena Math | 1400 | 1272 |
| MATH Level 5 | 96.6% | 46.7% |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
Knowledge DeepSeek-R1 leads
DeepSeek-R1: 44.5 (#87), GPT-4 Turbo: 24.3 (#268)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| GPQA Diamond | 76.3% | 46.6% |
| Confabulations | 12.7% | 28.4% |
| LMArena Expert | 1394 | 1223 |
| MMLU-Pro | 79.3% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |
| MMLU | — | 81.3% |
Multimodal Not comparable
DeepSeek-R1: —, GPT-4 Turbo: 30.6 (#110)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| LMArena Vision | — | 1090 |
Multilingual DeepSeek-R1 leads
DeepSeek-R1: 52.4 (#85), GPT-4 Turbo: 40.5 (#216)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| LMArena Non-English | 1412 | 1245 |
| LMArena Chinese | 1442 | 1242 |
| LMArena French | 1417 | 1276 |
| LMArena German | 1404 | 1259 |
| LMArena Japanese | 1391 | 1194 |
| LMArena Korean | 1360 | 1187 |
| LMArena Russian | 1423 | 1259 |
| LMArena Spanish | 1411 | 1260 |
Instruction Following DeepSeek-R1 leads
DeepSeek-R1: 72.0 (#143), GPT-4 Turbo: 65.8 (#216)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| LMArena Instruction Following | 1382 | 1249 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), GPT-4 Turbo: 38.0 (#206)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| LMArena Longer Query | 1391 | 1254 |
| Fiction.LiveBench | 75% | — |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), GPT-4 Turbo: 47.7 (#206)
| Benchmark | DeepSeek-R1 | GPT-4 Turbo |
|---|---|---|
| LMArena Text | 1428 | 1272 |
| LMArena Creative Writing | 1405 | 1269 |
| LMArena Multi-Turn | 1405 | 1267 |
| 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-4 Turbo?
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 30.5 on the Noometry Index.
Which is cheaper, DeepSeek-R1 or GPT-4 Turbo?
DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is DeepSeek-R1 or GPT-4 Turbo better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-R1 does, with 164K tokens against 128K.
How many benchmarks do DeepSeek-R1 and GPT-4 Turbo share?
26 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and GPT-4 Turbo has 36.