Model comparison
GPT-4.5 vs Grok 4.20 (Non-Reasoning)
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 37.2 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. GPT-4.5 scores higher in 2 categories and Grok 4.20 (Non-Reasoning) in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 13.9.
- The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 89.5% for Grok 4.20 (Non-Reasoning).
Side by side
| GPT-4.5 | Grok 4.20 (Non-Reasoning) | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 37.2 | 48.6 |
| Released | 2025-02-27 | 2026-02-17 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1M |
| Max output | — | 30K |
| Input $ / M tokens | — | $1.25 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 42 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-4.5: 42.2 (#109), Grok 4.20 (Non-Reasoning): 42.1 (#112)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| WeirdML | 39.4% | 52.3% |
| LMArena Coding | 1396 | 1459 |
| Aider Polyglot | 44.9% | — |
| LMArena WebDev | — | 1375 |
| LiveBench Coding | 75.2% | — |
| ALE-Bench | — | 1,150 |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
GPT-4.5: 27.9 (#97), Grok 4.20 (Non-Reasoning): 34.4 (#46)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Terminal-Bench | — | 57.3% |
| τ²-bench Banking | — | 18% |
| Cybench | 17.5% | — |
| LMArena Search | — | 1189 |
| Vending-Bench 2 | — | 4,663 |
Reasoning Grok 4.20 (Non-Reasoning) leads
GPT-4.5: 13.9 (#330), Grok 4.20 (Non-Reasoning): 52.3 (#32)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 0.8% | 65.1% |
| ARC-AGI-1 | 10.3% | 89.5% |
| LMArena Hard Prompts | 1403 | 1451 |
| Epoch Capabilities Index | 136.74 | 151.98 |
| ForecastBench | 61.7 | 61.4 |
| SimpleBench | 34.5% | — |
| Kagi LLM Benchmark | — | 75% |
| NYT Connections (extended) | — | 85.4% |
| Chess Puzzles | — | 24% |
| EnigmaEval | 3.2% | — |
| Thematic Generalization | — | 63.8% |
| LiveBench Reasoning | 71.1% | — |
| DTBench | — | 90.1% |
| LiveBench Data Analysis | 64.3% | — |
| LMCA | — | 38.7% |
| LiveBench | 69% | — |
Math Grok 4.20 (Non-Reasoning) leads
GPT-4.5: 32.6 (#211), Grok 4.20 (Non-Reasoning): 48.2 (#65)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 92.2% |
| LMArena Math | 1412 | 1455 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 14% |
| LiveBench Math | 69.3% | — |
| MATH Level 5 | 78.6% | — |
Knowledge Grok 4.20 (Non-Reasoning) leads
GPT-4.5: 32.5 (#211), Grok 4.20 (Non-Reasoning): 52.8 (#60)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 68.7% | 89.3% |
| LMArena Expert | 1394 | 1439 |
| Humanity's Last Exam | 5.4% | — |
| SimpleQA Verified | — | 30.2% |
| Confabulations | 13.6% | — |
Multimodal GPT-4.5 leads
GPT-4.5: 37.6 (#71), Grok 4.20 (Non-Reasoning): 33.3 (#98)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Vision | 1195 | 1263 |
| VPCT | 45% | — |
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1416 |
Multilingual Grok 4.20 (Non-Reasoning) leads
GPT-4.5: 52.5 (#83), Grok 4.20 (Non-Reasoning): 54.5 (#40)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1413 | 1441 |
| LMArena Chinese | 1421 | 1481 |
| LMArena French | 1418 | 1476 |
| LMArena German | 1457 | 1465 |
| LMArena Japanese | 1416 | 1449 |
| LMArena Korean | 1392 | 1417 |
| LMArena Russian | 1419 | 1458 |
| LMArena Spanish | — | 1443 |
Instruction Following Grok 4.20 (Non-Reasoning) leads
GPT-4.5: 72.6 (#134), Grok 4.20 (Non-Reasoning): 74.8 (#83)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1404 | 1420 |
| LiveBench Instruction Following | 72.3% | — |
Long Context Grok 4.20 (Non-Reasoning) leads
GPT-4.5: 40.4 (#155), Grok 4.20 (Non-Reasoning): 45.5 (#34)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Longer Query | 1406 | 1437 |
| Fiction.LiveBench | 63.9% | — |
| CL-bench | — | 22.2% |
| CL-bench Life | — | 11.9% |
Writing & Preference Grok 4.20 (Non-Reasoning) leads
GPT-4.5: 56.9 (#134), Grok 4.20 (Non-Reasoning): 65.7 (#44)
| Benchmark | GPT-4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1417 | 1451 |
| LMArena Creative Writing | 1394 | 1438 |
| EQ-Bench Creative Writing | 1258 | 1574 |
| LMArena Multi-Turn | 1444 | 1456 |
| Short-Story Creative Writing | 75.6% | — |
| LiveBench Language | 61.5% | — |
Frequently asked questions
Is GPT-4.5 better than Grok 4.20 (Non-Reasoning)?
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 37.2 on the Noometry Index.
Is GPT-4.5 or Grok 4.20 (Non-Reasoning) better for coding?
They score almost the same on coding (42.2 vs 42.1); test both on your own repository before choosing.
How many benchmarks do GPT-4.5 and Grok 4.20 (Non-Reasoning) share?
25 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.