Model comparison
DeepSeek-R1 vs GPT-5 Mini
DeepSeek-R1 and GPT-5 Mini score almost the same on the Noometry Index (42.3 vs 41.8), so choose on price, context window or the category you care about most.
Last verified . 40 shared benchmarks.
DeepSeekSummary
- They share 40 benchmarks with published results for both. DeepSeek-R1 scores higher in 4 categories and GPT-5 Mini in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-R1 leads 46.3 to 40.1.
- The biggest single-benchmark swing is ARC-AGI-1: 21.2% for DeepSeek-R1 and 54.3% for GPT-5 Mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $0.50 / $2.15 for DeepSeek-R1.
- GPT-5 Mini accepts more context: 400K tokens versus 164K.
Side by side
| DeepSeek-R1 | GPT-5 Mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 42.3 | 41.8 |
| Released | 2025-01-20 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 164K | 400K |
| Max output | 64K | 128K |
| Input $ / M tokens | $0.50 | $0.25 |
| Output $ / M tokens | $2.15 | $2 |
| 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), GPT-5 Mini: 40.1 (#146)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| SciCode | 35.7% | 39.2% |
| WeirdML | 41.6% | 52.7% |
| LMArena Coding | 1427 | 1406 |
| ALE-Bench | 804.12 | 799.77 |
| AlgoTune | 1.7 | 1.38 |
| SWE-bench Verified | — | 64.7% |
| SWE-bench Verified (bash only) | — | 59.8% |
| Aider Polyglot | 71.4% | — |
| SWE-bench Multilingual | — | 39.7% |
| LiveBench Coding | 66.7% | — |
Agentic & Tool Use Too close to call
DeepSeek-R1: 30.7 (#75), GPT-5 Mini: 31.1 (#70)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
| Vending-Bench 2 | — | -31.18 |
Reasoning GPT-5 Mini leads
DeepSeek-R1: 18.6 (#278), GPT-5 Mini: 23.9 (#168)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 1.3% | 4.4% |
| Kagi LLM Benchmark | 69.4% | 70.3% |
| ARC-AGI-1 | 21.2% | 54.3% |
| CritPt | 1.1% | 0% |
| LMArena Hard Prompts | 1416 | 1380 |
| Epoch Capabilities Index | 141.29 | 145.52 |
| ForecastBench | 60 | 61 |
| SimpleBench | 40.8% | — |
| Chess Puzzles | — | 30% |
| EnigmaEval | — | 8.2% |
| LiveBench Reasoning | 83.2% | — |
| Mystery Game Puzzles | — | 10% |
| DTBench | — | 80.5% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 34.2% |
| LiveBench | 71.6% | — |
Math GPT-5 Mini leads
DeepSeek-R1: 43.8 (#79), GPT-5 Mini: 46.7 (#69)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 86.7% |
| Omni-MATH | 42.4% | 72.2% |
| LMArena Math | 1400 | 1378 |
| MATH Level 5 | 96.6% | 97.8% |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 9% |
| LiveBench Math | 80.7% | — |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5 Mini leads
DeepSeek-R1: 44.5 (#87), GPT-5 Mini: 45.6 (#86)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 76.3% | 75% |
| MMLU-Pro | 79.3% | 83.5% |
| Confabulations | 12.7% | 13.3% |
| Vectara Hallucination Rate | 11.3% | 12.9% |
| GPQA (HELM) | 66.6% | 75.6% |
| LMArena Expert | 1394 | 1379 |
| Humanity's Last Exam | — | 19.4% |
| SimpleQA Verified | — | 21.6% |
Multimodal Not comparable
DeepSeek-R1: —, GPT-5 Mini: 35.6 (#85)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| LMArena Vision | — | 1202 |
| VPCT | — | 40.2% |
Multilingual DeepSeek-R1 leads
DeepSeek-R1: 52.4 (#85), GPT-5 Mini: 48.9 (#137)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1412 | 1363 |
| LMArena Chinese | 1442 | 1385 |
| LMArena French | 1417 | 1386 |
| LMArena German | 1404 | 1366 |
| LMArena Japanese | 1391 | 1341 |
| LMArena Korean | 1360 | 1308 |
| LMArena Russian | 1423 | 1362 |
| LMArena Spanish | 1411 | 1355 |
Instruction Following GPT-5 Mini leads
DeepSeek-R1: 72.0 (#143), GPT-5 Mini: 76.2 (#46)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| IFEval | 78.4% | 92.7% |
| LMArena Instruction Following | 1382 | 1357 |
| LiveBench Instruction Following | 80.5% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), GPT-5 Mini: 41.9 (#132)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| Fiction.LiveBench | 75% | 69.4% |
| LMArena Longer Query | 1391 | 1355 |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), GPT-5 Mini: 55.2 (#148)
| Benchmark | DeepSeek-R1 | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1428 | 1373 |
| LMArena Creative Writing | 1405 | 1325 |
| Short-Story Creative Writing | 83% | 83.1% |
| EQ-Bench Creative Writing | 1500 | 1313 |
| WildBench | 82.8% | 85.5% |
| LMArena Multi-Turn | 1405 | 1363 |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than GPT-5 Mini?
DeepSeek-R1 and GPT-5 Mini score almost the same on the Noometry Index (42.3 vs 41.8), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-R1 or GPT-5 Mini?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; DeepSeek-R1 lists at $0.50 and $2.15.
Is DeepSeek-R1 or GPT-5 Mini better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 164K.
How many benchmarks do DeepSeek-R1 and GPT-5 Mini share?
40 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and GPT-5 Mini has 60.