Model comparison
DeepSeek-R1 vs o4-mini
DeepSeek-R1 and o4-mini score almost the same on the Noometry Index (42.3 vs 41.6), so choose on price, context window or the category you care about most.
Last verified . 41 shared benchmarks.
DeepSeekSummary
- They share 41 benchmarks with published results for both. DeepSeek-R1 scores higher in 5 categories and o4-mini in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-R1 leads 61.4 to 54.0.
- The biggest single-benchmark swing is ARC-AGI-1: 21.2% for DeepSeek-R1 and 58.7% for o4-mini.
- DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $1.10 / $4.40 for o4-mini.
- o4-mini accepts more context: 200K tokens versus 164K.
Side by side
| DeepSeek-R1 | o4-mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 42.3 | 41.6 |
| Released | 2025-01-20 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | 164K | 200K |
| Max output | 64K | 100K |
| Input $ / M tokens | $0.50 | $1.10 |
| Output $ / M tokens | $2.15 | $4.40 |
| Results tracked | 52 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), o4-mini: 40.9 (#127)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| Aider Polyglot | 71.4% | 72% |
| WeirdML | 41.6% | 52.6% |
| LMArena Coding | 1427 | 1368 |
| ALE-Bench | 804.12 | 826.17 |
| AlgoTune | 1.7 | 1.72 |
| SWE-bench Verified (bash only) | — | 45% |
| SciCode | 35.7% | — |
| GSO | — | 3.6% |
| LiveBench Coding | 66.7% | — |
| CadEval | — | 62% |
Agentic & Tool Use o4-mini leads
DeepSeek-R1: 30.7 (#75), o4-mini: 32.6 (#61)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| METR Time Horizons | 53.8% | 63.9% |
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
Reasoning o4-mini leads
DeepSeek-R1: 18.6 (#278), o4-mini: 24.6 (#162)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| ARC-AGI-2 | 1.3% | 6.1% |
| SimpleBench | 40.8% | 38.7% |
| Kagi LLM Benchmark | 69.4% | 67.6% |
| ARC-AGI-1 | 21.2% | 58.7% |
| CritPt | 1.1% | 0.6% |
| LMArena Hard Prompts | 1416 | 1351 |
| Epoch Capabilities Index | 141.29 | 145.64 |
| ForecastBench | 60 | 61.8 |
| Chess Puzzles | — | 26% |
| EnigmaEval | — | 9.2% |
| LiveBench Reasoning | 83.2% | — |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 77.6% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 26.5% |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), o4-mini: 40.8 (#89)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 81.7% |
| Omni-MATH | 42.4% | 72% |
| LMArena Math | 1400 | 1389 |
| MATH Level 5 | 96.6% | 97.8% |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| LiveBench Math | 80.7% | — |
| FrontierMath (Feb 2025 set) | — | 24.8% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge Too close to call
DeepSeek-R1: 44.5 (#87), o4-mini: 43.6 (#91)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| GPQA Diamond | 76.3% | 79.6% |
| MMLU-Pro | 79.3% | 82% |
| Confabulations | 12.7% | 15.8% |
| Vectara Hallucination Rate | 11.3% | 18.6% |
| GPQA (HELM) | 66.6% | 73.5% |
| LMArena Expert | 1394 | 1343 |
| Humanity's Last Exam | — | 18.1% |
| SimpleQA Verified | — | 19.6% |
Multimodal Not comparable
DeepSeek-R1: —, o4-mini: 40.2 (#49)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| LMArena Vision | — | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |
Multilingual DeepSeek-R1 leads
DeepSeek-R1: 52.4 (#85), o4-mini: 47.0 (#154)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| LMArena Non-English | 1412 | 1337 |
| LMArena Chinese | 1442 | 1354 |
| LMArena French | 1417 | 1364 |
| LMArena German | 1404 | 1336 |
| LMArena Japanese | 1391 | 1308 |
| LMArena Korean | 1360 | 1312 |
| LMArena Russian | 1423 | 1334 |
| LMArena Spanish | 1411 | 1347 |
Instruction Following o4-mini leads
DeepSeek-R1: 72.0 (#143), o4-mini: 75.2 (#68)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| IFEval | 78.4% | 92.8% |
| LMArena Instruction Following | 1382 | 1321 |
| LiveBench Instruction Following | 80.5% | — |
Long Context Too close to call
DeepSeek-R1: 45.4 (#36), o4-mini: 45.5 (#33)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| Fiction.LiveBench | 75% | 77.8% |
| LMArena Longer Query | 1391 | 1315 |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), o4-mini: 54.0 (#152)
| Benchmark | DeepSeek-R1 | o4-mini |
|---|---|---|
| LMArena Text | 1428 | 1353 |
| LMArena Creative Writing | 1405 | 1294 |
| Short-Story Creative Writing | 83% | 75% |
| WildBench | 82.8% | 85.4% |
| LMArena Multi-Turn | 1405 | 1350 |
| EQ-Bench Creative Writing | 1500 | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than o4-mini?
DeepSeek-R1 and o4-mini score almost the same on the Noometry Index (42.3 vs 41.6), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-R1 or o4-mini?
DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; o4-mini lists at $1.10 and $4.40.
Is DeepSeek-R1 or o4-mini better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 40.9 in the Noometry coding category.
Which has the bigger context window?
o4-mini does, with 200K tokens against 164K.
How many benchmarks do DeepSeek-R1 and o4-mini share?
41 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and o4-mini has 60.