Model comparison
DeepSeek-R1 vs Grok 4.6
Grok 4.6 is the stronger model overall, scoring 56.9 to 42.3 on the Noometry Index. DeepSeek-R1 costs 3.3× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
DeepSeekSummary
- They share 27 benchmarks with published results for both. DeepSeek-R1 scores higher in 1 category and Grok 4.6 in 8 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 18.6.
- The biggest single-benchmark swing is ARC-AGI-1: 21.2% for DeepSeek-R1 and 87.5% for Grok 4.6.
- DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 164K.
Side by side
| DeepSeek-R1 | Grok 4.6 | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 42.3 | 56.9 |
| Released | 2025-01-20 | 2026-08-12 |
| Weights | Proprietary | Proprietary |
| Context window | 164K | 500K |
| Max output | 64K | 500K |
| Input $ / M tokens | $0.50 | $2 |
| Output $ / M tokens | $2.15 | $6 |
| Results tracked | 52 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
DeepSeek-R1: 46.3 (#68), Grok 4.6: 58.5 (#16)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| SciCode | 35.7% | 56.5% |
| WeirdML | 41.6% | 67.3% |
| LMArena Coding | 1427 | 1465 |
| ALE-Bench | 804.12 | 1,508 |
| DeepSWE | — | 67.5% |
| FrontierCode | — | 48% |
| Aider Polyglot | 71.4% | — |
| CursorBench | — | 41.4% |
| LMArena WebDev | — | 1617 |
| FrontierSWE | — | 25.3% |
| LiveBench Coding | 66.7% | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use Grok 4.6 leads
DeepSeek-R1: 30.7 (#75), Grok 4.6: 39.4 (#27)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| APEX-Agents | — | 65.3% |
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| GDP.pdf | — | 17.2% |
| METR Time Horizons | 53.8% | — |
| Vending-Bench 2 | — | 9,047 |
Reasoning Grok 4.6 leads
DeepSeek-R1: 18.6 (#278), Grok 4.6: 61.4 (#20)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| ARC-AGI-2 | 1.3% | 67.1% |
| SimpleBench | 40.8% | 75.9% |
| ARC-AGI-1 | 21.2% | 87.5% |
| CritPt | 1.1% | 19.7% |
| LMArena Hard Prompts | 1416 | 1447 |
| Epoch Capabilities Index | 141.29 | 156.44 |
| Kagi LLM Benchmark | 69.4% | — |
| NYT Connections (extended) | — | 80% |
| Chess Puzzles | — | 40% |
| EBR-Bench | — | 30.5% |
| LiveBench Reasoning | 83.2% | — |
| Mystery Game Puzzles | — | 34% |
| DTBench | — | 97.3% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 48.5% |
| ForecastBench | 60 | — |
| LiveBench | 71.6% | — |
Math Grok 4.6 leads
DeepSeek-R1: 43.8 (#79), Grok 4.6: 67.0 (#24)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 99.2% |
| LMArena Math | 1400 | 1423 |
| FrontierMath (Tiers 1-3) | — | 66% |
| FrontierMath Tier 4 | — | 31.7% |
| ProofBench | — | 51% |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |
Knowledge Grok 4.6 leads
DeepSeek-R1: 44.5 (#87), Grok 4.6: 63.3 (#20)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 76.3% | 94% |
| LMArena Expert | 1394 | 1467 |
| SimpleQA Verified | — | 49.3% |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |
Multimodal Not comparable
DeepSeek-R1: —, Grok 4.6: 43.6 (#23)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| LMArena Vision | — | 1263 |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| LMArena Document | — | 1452 |
Multilingual Too close to call
DeepSeek-R1: 52.4 (#85), Grok 4.6: 53.0 (#74)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1412 | 1420 |
| LMArena Chinese | 1442 | 1480 |
| LMArena French | 1417 | 1461 |
| LMArena German | 1404 | 1431 |
| LMArena Japanese | 1391 | 1376 |
| LMArena Korean | 1360 | 1397 |
| LMArena Russian | 1423 | 1422 |
| LMArena Spanish | 1411 | 1404 |
Instruction Following Grok 4.6 leads
DeepSeek-R1: 72.0 (#143), Grok 4.6: 75.4 (#63)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1382 | 1431 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context Too close to call
DeepSeek-R1: 45.4 (#36), Grok 4.6: 44.5 (#66)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1391 | 1454 |
| Fiction.LiveBench | 75% | — |
Writing & Preference Too close to call
DeepSeek-R1: 61.4 (#88), Grok 4.6: 62.3 (#80)
| Benchmark | DeepSeek-R1 | Grok 4.6 |
|---|---|---|
| LMArena Text | 1428 | 1428 |
| LMArena Creative Writing | 1405 | 1428 |
| LMArena Multi-Turn | 1405 | 1425 |
| 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 Grok 4.6?
Grok 4.6 is the stronger model overall, scoring 56.9 to 42.3 on the Noometry Index. DeepSeek-R1 costs 3.3× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-R1 or Grok 4.6?
DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Grok 4.6 lists at $2 and $6.
Is DeepSeek-R1 or Grok 4.6 better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 46.3 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 164K.
How many benchmarks do DeepSeek-R1 and Grok 4.6 share?
27 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Grok 4.6 has 49.