Model comparison
DeepSeek-R1 vs Grok 4.20 Multi-Agent
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 42.3 on the Noometry Index. DeepSeek-R1 costs 1.7× less per token, which makes it the better buy when Grok 4.20 Multi-Agent's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. DeepSeek-R1 scores higher in 4 categories and Grok 4.20 Multi-Agent in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 Multi-Agent leads 43.9 to 18.6.
- DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 Multi-Agent.
- Grok 4.20 Multi-Agent accepts more context: 1M tokens versus 164K.
Side by side
| DeepSeek-R1 | Grok 4.20 Multi-Agent | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 42.3 | 46.2 |
| Released | 2025-01-20 | 2026-03-09 |
| Weights | Proprietary | Proprietary |
| Context window | 164K | 1M |
| Max output | 64K | 30K |
| Input $ / M tokens | $0.50 | $1.25 |
| Output $ / M tokens | $2.15 | $2.50 |
| Results tracked | 52 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), Grok 4.20 Multi-Agent: 43.0 (#92)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Coding | 1427 | 1457 |
| Aider Polyglot | 71.4% | — |
| SciCode | 35.7% | — |
| WeirdML | 41.6% | — |
| LiveBench Coding | 66.7% | — |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use Not comparable
DeepSeek-R1: 30.7 (#75), Grok 4.20 Multi-Agent: —
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| LMArena Search | — | 1204 |
| METR Time Horizons | 53.8% | — |
Reasoning Grok 4.20 Multi-Agent leads
DeepSeek-R1: 18.6 (#278), Grok 4.20 Multi-Agent: 43.9 (#48)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Hard Prompts | 1416 | 1448 |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 40.8% | — |
| Kagi LLM Benchmark | 69.4% | — |
| NYT Connections (extended) | — | 89.6% |
| ARC-AGI-1 | 21.2% | — |
| CritPt | 1.1% | — |
| LiveBench Reasoning | 83.2% | — |
| LiveBench Data Analysis | 69.8% | — |
| Epoch Capabilities Index | 141.29 | — |
| ForecastBench | 60 | — |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), Grok 4.20 Multi-Agent: 39.4 (#104)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Math | 1400 | 1442 |
| OTIS Mock AIME 2024-2025 | 66.4% | — |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |
Knowledge DeepSeek-R1 leads
DeepSeek-R1: 44.5 (#87), Grok 4.20 Multi-Agent: 40.4 (#119)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Expert | 1394 | 1445 |
| GPQA Diamond | 76.3% | — |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |
Multimodal Not comparable
DeepSeek-R1: —, Grok 4.20 Multi-Agent: 40.5 (#48)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Vision | — | 1259 |
Multilingual Grok 4.20 Multi-Agent leads
DeepSeek-R1: 52.4 (#85), Grok 4.20 Multi-Agent: 54.4 (#43)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Non-English | 1412 | 1440 |
| LMArena Chinese | 1442 | 1475 |
| LMArena French | 1417 | 1466 |
| LMArena German | 1404 | 1456 |
| LMArena Japanese | 1391 | 1405 |
| LMArena Korean | 1360 | 1416 |
| LMArena Russian | 1423 | 1457 |
| LMArena Spanish | 1411 | 1447 |
Instruction Following Grok 4.20 Multi-Agent leads
DeepSeek-R1: 72.0 (#143), Grok 4.20 Multi-Agent: 74.8 (#84)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Instruction Following | 1382 | 1420 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), Grok 4.20 Multi-Agent: 43.7 (#88)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Longer Query | 1391 | 1431 |
| Fiction.LiveBench | 75% | — |
Writing & Preference Grok 4.20 Multi-Agent leads
DeepSeek-R1: 61.4 (#88), Grok 4.20 Multi-Agent: 64.0 (#59)
| Benchmark | DeepSeek-R1 | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Text | 1428 | 1450 |
| LMArena Creative Writing | 1405 | 1436 |
| LMArena Multi-Turn | 1405 | 1452 |
| 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 4.20 Multi-Agent?
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 42.3 on the Noometry Index. DeepSeek-R1 costs 1.7× less per token, which makes it the better buy when Grok 4.20 Multi-Agent's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-R1 or Grok 4.20 Multi-Agent?
DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Grok 4.20 Multi-Agent lists at $1.25 and $2.50.
Is DeepSeek-R1 or Grok 4.20 Multi-Agent better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 43.0 in the Noometry coding category.
Which has the bigger context window?
Grok 4.20 Multi-Agent does, with 1M tokens against 164K.
How many benchmarks do DeepSeek-R1 and Grok 4.20 Multi-Agent share?
17 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Grok 4.20 Multi-Agent has 20.