Model comparison
Grok 4 vs Grok 4.20 (Non-Reasoning)
Grok 4 and Grok 4.20 (Non-Reasoning) score almost the same on the Noometry Index (48.1 vs 48.6), so choose on price, context window or the category you care about most.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. Grok 4 scores higher in 6 categories and Grok 4.20 (Non-Reasoning) in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 45.5.
- The biggest single-benchmark swing is ARC-AGI-2: 16% for Grok 4 and 65.1% for Grok 4.20 (Non-Reasoning).
Side by side
| Grok 4 | Grok 4.20 (Non-Reasoning) | |
|---|---|---|
| Provider | xAI | xAI |
| Noometry Index | 48.1 | 48.6 |
| Released | 2025-07-09 | 2026-02-17 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1M |
| Max output | — | 30K |
| Input $ / M tokens | — | $1.25 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 48 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4 leads
Grok 4: 50.3 (#46), Grok 4.20 (Non-Reasoning): 42.1 (#112)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| WeirdML | 45.7% | 52.3% |
| LMArena Coding | 1408 | 1459 |
| Aider Polyglot | 79.6% | — |
| LMArena WebDev | — | 1375 |
| ALE-Bench | — | 1,150 |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
Grok 4: 32.3 (#68), Grok 4.20 (Non-Reasoning): 34.4 (#46)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Terminal-Bench | 27.2% | 57.3% |
| LMArena Search | 1142 | 1189 |
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 21.1% | — |
| τ²-bench Banking | — | 18% |
| Cybench | 43% | — |
| DeepResearch Bench | 47.3% | — |
| BALROG | 43.6% | — |
| METR Time Horizons | 66.6% | — |
| Vending-Bench 2 | — | 4,663 |
Reasoning Grok 4.20 (Non-Reasoning) leads
Grok 4: 36.7 (#65), Grok 4.20 (Non-Reasoning): 52.3 (#32)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 16% | 65.1% |
| Kagi LLM Benchmark | 73.6% | 75% |
| ARC-AGI-1 | 66.7% | 89.5% |
| Chess Puzzles | 28% | 24% |
| LMArena Hard Prompts | 1409 | 1451 |
| Epoch Capabilities Index | 146.44 | 151.98 |
| ForecastBench | 60.9 | 61.4 |
| SimpleBench | 60.5% | — |
| NYT Connections (extended) | — | 85.4% |
| Thematic Generalization | — | 63.8% |
| DTBench | — | 90.1% |
| LMCA | — | 38.7% |
Math Too close to call
Grok 4: 48.4 (#64), Grok 4.20 (Non-Reasoning): 48.2 (#65)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 84% | 92.2% |
| LMArena Math | 1422 | 1455 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 14% |
| Omni-MATH | 60.3% | — |
| FrontierMath (Feb 2025 set) | 19.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Too close to call
Grok 4: 53.8 (#55), Grok 4.20 (Non-Reasoning): 52.8 (#60)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 87% | 89.3% |
| LMArena Expert | 1415 | 1439 |
| SimpleQA Verified | — | 30.2% |
| MMLU-Pro | 85.1% | — |
| Confabulations | 12.4% | — |
| GPQA (HELM) | 72.7% | — |
Multimodal Too close to call
Grok 4: 33.7 (#94), Grok 4.20 (Non-Reasoning): 33.3 (#98)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Vision | 1210 | 1263 |
| GeoBench | 45% | — |
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1416 |
Multilingual Grok 4.20 (Non-Reasoning) leads
Grok 4: 51.8 (#103), Grok 4.20 (Non-Reasoning): 54.5 (#40)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1403 | 1441 |
| LMArena Chinese | 1427 | 1481 |
| LMArena French | 1418 | 1476 |
| LMArena German | 1429 | 1465 |
| LMArena Japanese | 1394 | 1449 |
| LMArena Korean | 1377 | 1417 |
| LMArena Russian | 1410 | 1458 |
| LMArena Spanish | 1420 | 1443 |
Instruction Following Grok 4 leads
Grok 4: 79.2 (#5), Grok 4.20 (Non-Reasoning): 74.8 (#83)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1387 | 1420 |
| IFEval | 94.9% | — |
Long Context Grok 4 leads
Grok 4: 63.1 (#4), Grok 4.20 (Non-Reasoning): 45.5 (#34)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Longer Query | 1409 | 1437 |
| Fiction.LiveBench | 94.4% | — |
| CL-bench | — | 22.2% |
| CL-bench Life | — | 11.9% |
Writing & Preference Grok 4.20 (Non-Reasoning) leads
Grok 4: 58.5 (#116), Grok 4.20 (Non-Reasoning): 65.7 (#44)
| Benchmark | Grok 4 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1411 | 1451 |
| LMArena Creative Writing | 1397 | 1438 |
| LMArena Multi-Turn | 1416 | 1456 |
| Short-Story Creative Writing | 76.9% | — |
| EQ-Bench Creative Writing | — | 1574 |
| WildBench | 79.7% | — |
Frequently asked questions
Is Grok 4 better than Grok 4.20 (Non-Reasoning)?
Grok 4 and Grok 4.20 (Non-Reasoning) score almost the same on the Noometry Index (48.1 vs 48.6), so choose on price, context window or the category you care about most.
Is Grok 4 or Grok 4.20 (Non-Reasoning) better for coding?
Grok 4 scores higher on coding benchmarks: 50.3 versus 42.1 in the Noometry coding category.
How many benchmarks do Grok 4 and Grok 4.20 (Non-Reasoning) share?
29 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.