Model comparison
DeepSeek-R1 vs DeepSeek-V3.1
DeepSeek-R1 and DeepSeek-V3.1 score almost the same on the Noometry Index (42.3 vs 42.8), so choose on price, context window or the category you care about most.
Last verified . 25 shared benchmarks.
DeepSeekSummary
- They share 25 benchmarks with published results for both. DeepSeek-R1 scores higher in 6 categories and DeepSeek-V3.1 in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V3.1 leads 27.9 to 18.6.
- The biggest single-benchmark swing is Fiction.LiveBench: 75% for DeepSeek-R1 and 52.8% for DeepSeek-V3.1.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.50 / $2.15 for DeepSeek-R1.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1 | DeepSeek-V3.1 | |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 42.3 | 42.8 |
| Released | 2025-01-20 | 2025-08-21 |
| Weights | Proprietary | Open |
| Context window | 164K | 164K |
| Max output | 64K | 8K |
| Input $ / M tokens | $0.50 | $0.25 |
| Output $ / M tokens | $2.15 | $0.95 |
| Results tracked | 52 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), DeepSeek-V3.1: 40.3 (#144)
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| WeirdML | 41.6% | 38.4% |
| LMArena Coding | 1427 | 1417 |
| Aider Polyglot | 71.4% | — |
| SciCode | 35.7% | — |
| LiveBench Coding | 66.7% | — |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use Not comparable
DeepSeek-R1: 30.7 (#75), DeepSeek-V3.1: —
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
Reasoning DeepSeek-V3.1 leads
DeepSeek-R1: 18.6 (#278), DeepSeek-V3.1: 27.9 (#110)
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| SimpleBench | 40.8% | 40% |
| Kagi LLM Benchmark | 69.4% | 53.2% |
| LMArena Hard Prompts | 1416 | 1417 |
| Epoch Capabilities Index | 141.29 | 139.92 |
| ForecastBench | 60 | 58 |
| ARC-AGI-2 | 1.3% | — |
| ARC-AGI-1 | 21.2% | — |
| CritPt | 1.1% | — |
| LiveBench Reasoning | 83.2% | — |
| DTBench | — | 82.7% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 24.3% |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), DeepSeek-V3.1: 38.9 (#122)
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Math | 1400 | 1420 |
| OTIS Mock AIME 2024-2025 | 66.4% | — |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |
Knowledge Too close to call
DeepSeek-R1: 44.5 (#87), DeepSeek-V3.1: 43.7 (#90)
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| Vectara Hallucination Rate | 11.3% | 5.5% |
| LMArena Expert | 1394 | 1405 |
| GPQA Diamond | 76.3% | — |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| GPQA (HELM) | 66.6% | — |
Multilingual Too close to call
DeepSeek-R1: 52.4 (#85), DeepSeek-V3.1: 51.6 (#106)
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Non-English | 1412 | 1400 |
| LMArena Chinese | 1442 | 1469 |
| LMArena French | 1417 | 1447 |
| LMArena German | 1404 | 1411 |
| LMArena Japanese | 1391 | 1378 |
| LMArena Korean | 1360 | 1337 |
| LMArena Russian | 1423 | 1405 |
| LMArena Spanish | 1411 | 1431 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-R1: 72.0 (#143), DeepSeek-V3.1: 73.9 (#110)
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Instruction Following | 1382 | 1400 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), DeepSeek-V3.1: 36.3 (#232)
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| Fiction.LiveBench | 75% | 52.8% |
| LMArena Longer Query | 1391 | 1422 |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), DeepSeek-V3.1: 60.3 (#98)
| Benchmark | DeepSeek-R1 | DeepSeek-V3.1 |
|---|---|---|
| LMArena Text | 1428 | 1420 |
| LMArena Creative Writing | 1405 | 1401 |
| EQ-Bench Creative Writing | 1500 | 1436 |
| LMArena Multi-Turn | 1405 | 1408 |
| Short-Story Creative Writing | 83% | — |
| WildBench | 82.8% | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than DeepSeek-V3.1?
DeepSeek-R1 and DeepSeek-V3.1 score almost the same on the Noometry Index (42.3 vs 42.8), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-R1 or DeepSeek-V3.1?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; DeepSeek-R1 lists at $0.50 and $2.15.
Is DeepSeek-R1 or DeepSeek-V3.1 better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 40.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 164K tokens.
How many benchmarks do DeepSeek-R1 and DeepSeek-V3.1 share?
25 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and DeepSeek-V3.1 has 27.