Model comparison
DeepSeek-R1 vs Grok 4.5
Grok 4.5 is the stronger model overall, scoring 55.0 to 42.3 on the Noometry Index. DeepSeek-R1 costs 3.3× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
DeepSeekSummary
- They share 29 benchmarks with published results for both. DeepSeek-R1 scores higher in 1 category and Grok 4.5 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.5 leads 56.1 to 18.6.
- The biggest single-benchmark swing is ARC-AGI-1: 21.2% for DeepSeek-R1 and 87.2% for Grok 4.5.
- DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $2 / $6 for Grok 4.5.
- Grok 4.5 accepts more context: 500K tokens versus 164K.
Side by side
| DeepSeek-R1 | Grok 4.5 | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 42.3 | 55.0 |
| Released | 2025-01-20 | 2026-07-08 |
| 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 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.5 leads
DeepSeek-R1: 46.3 (#68), Grok 4.5: 52.2 (#35)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| SciCode | 35.7% | 54.1% |
| WeirdML | 41.6% | 46.4% |
| LMArena Coding | 1427 | 1474 |
| ALE-Bench | 804.12 | 1,309 |
| DeepSWE | — | 53.8% |
| FrontierCode | — | 42.4% |
| Aider Polyglot | 71.4% | — |
| LMArena WebDev | — | 1553 |
| LiveBench Coding | 66.7% | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use Grok 4.5 leads
DeepSeek-R1: 30.7 (#75), Grok 4.5: 44.4 (#17)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| APEX-Agents | — | 56.2% |
| τ²-bench Banking | — | 47.9% |
| DeepResearch Bench | 35.1% | — |
| PostTrainBench | — | 23.4% |
| BALROG | 34.9% | — |
| GBAEval | — | 65.4% |
| GDP.pdf | — | 14% |
| LMArena Search | — | 1213 |
| METR Time Horizons | 53.8% | — |
| Vending-Bench 2 | — | 3,887 |
Reasoning Grok 4.5 leads
DeepSeek-R1: 18.6 (#278), Grok 4.5: 56.1 (#25)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| ARC-AGI-2 | 1.3% | 52.6% |
| SimpleBench | 40.8% | 70% |
| Kagi LLM Benchmark | 69.4% | 83.5% |
| ARC-AGI-1 | 21.2% | 87.2% |
| CritPt | 1.1% | 15.4% |
| LMArena Hard Prompts | 1416 | 1462 |
| Epoch Capabilities Index | 141.29 | 153.92 |
| NYT Connections (extended) | — | 79.9% |
| Chess Puzzles | — | 36% |
| LiveBench Reasoning | 83.2% | — |
| DTBench | — | 96.5% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 45.2% |
| Surface Evolver Bench | — | 74.4% |
| ForecastBench | 60 | — |
| LiveBench | 71.6% | — |
Math Grok 4.5 leads
DeepSeek-R1: 43.8 (#79), Grok 4.5: 60.9 (#35)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 97.8% |
| LMArena Math | 1400 | 1459 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 24.4% |
| ProofBench | — | 31% |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |
Knowledge Grok 4.5 leads
DeepSeek-R1: 44.5 (#87), Grok 4.5: 62.3 (#24)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| GPQA Diamond | 76.3% | 93.4% |
| LMArena Expert | 1394 | 1466 |
| SimpleQA Verified | — | 48.3% |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |
Multimodal Not comparable
DeepSeek-R1: —, Grok 4.5: 37.6 (#72)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| LMArena Vision | — | 1288 |
| Blueprint-Bench 2 | — | 27.3% |
| Furniture Assembly | — | 22.5% |
| LMArena Document | — | 1452 |
Multilingual Grok 4.5 leads
DeepSeek-R1: 52.4 (#85), Grok 4.5: 54.4 (#42)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| LMArena Non-English | 1412 | 1440 |
| LMArena Chinese | 1442 | 1496 |
| LMArena French | 1417 | 1456 |
| LMArena German | 1404 | 1446 |
| LMArena Japanese | 1391 | 1428 |
| LMArena Korean | 1360 | 1404 |
| LMArena Russian | 1423 | 1448 |
| LMArena Spanish | 1411 | 1450 |
Instruction Following Grok 4.5 leads
DeepSeek-R1: 72.0 (#143), Grok 4.5: 76.0 (#48)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| LMArena Instruction Following | 1382 | 1446 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context Too close to call
DeepSeek-R1: 45.4 (#36), Grok 4.5: 44.8 (#56)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| LMArena Longer Query | 1391 | 1463 |
| Fiction.LiveBench | 75% | — |
Writing & Preference Grok 4.5 leads
DeepSeek-R1: 61.4 (#88), Grok 4.5: 65.8 (#42)
| Benchmark | DeepSeek-R1 | Grok 4.5 |
|---|---|---|
| LMArena Text | 1428 | 1448 |
| LMArena Creative Writing | 1405 | 1442 |
| EQ-Bench Creative Writing | 1500 | 1579 |
| LMArena Multi-Turn | 1405 | 1456 |
| Short-Story Creative Writing | 83% | — |
| WildBench | 82.8% | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than Grok 4.5?
Grok 4.5 is the stronger model overall, scoring 55.0 to 42.3 on the Noometry Index. DeepSeek-R1 costs 3.3× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-R1 or Grok 4.5?
DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Grok 4.5 lists at $2 and $6.
Is DeepSeek-R1 or Grok 4.5 better for coding?
Grok 4.5 scores higher on coding benchmarks: 52.2 versus 46.3 in the Noometry coding category.
Which has the bigger context window?
Grok 4.5 does, with 500K tokens against 164K.
How many benchmarks do DeepSeek-R1 and Grok 4.5 share?
29 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Grok 4.5 has 52.