Model comparison
GPT-5 vs Grok 4.20 (Non-Reasoning)
GPT-5 is the stronger model overall, scoring 50.9 to 48.5 on the Noometry Index. Grok 4.20 (Non-Reasoning) costs 2.2× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.
Last verified . 38 shared benchmarks.
Summary
- They share 38 benchmarks with published results for both. GPT-5 scores higher in 5 categories and Grok 4.20 (Non-Reasoning) in 5 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in long context, where GPT-5 leads 69.5 to 45.4.
- The biggest single-benchmark swing is ARC-AGI-2: 9.9% for GPT-5 and 65.1% for Grok 4.20 (Non-Reasoning).
- Grok 4.20 (Non-Reasoning) is cheaper at $1.25 / $2.50 per million input/output tokens, against $1.25 / $10 for GPT-5.
- Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 400K.
Side by side
| GPT-5 | Grok 4.20 (Non-Reasoning) | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 50.9 | 48.5 |
| Released | 2025-08-07 | 2026-02-17 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1M |
| Max output | 128K | 30K |
| Input $ / M tokens | $1.25 | $1.25 |
| Output $ / M tokens | $10 | $2.50 |
| Results tracked | 69 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 leads
GPT-5: 50.7 (#39), Grok 4.20 (Non-Reasoning): 41.8 (#104)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena WebDev | 1417 | 1374 |
| WeirdML | 60.7% | 52.3% |
| LMArena Coding | 1436 | 1458 |
| ALE-Bench | 1,162 | 1,150 |
| SWE-bench Verified | 73.6% | — |
| SWE-bench Verified (bash only) | 65% | — |
| Aider Polyglot | 88% | — |
| SciCode | 42.9% | — |
| GSO | 6.9% | — |
| AlgoTune | 1.67 | — |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
GPT-5: 33.0 (#53), Grok 4.20 (Non-Reasoning): 34.3 (#43)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Terminal-Bench | 49.6% | 57.3% |
| LMArena Search | 1133 | 1189 |
| GDPval | 34.8% | — |
| Remote Labor Index | 1.7% | — |
| τ²-bench Banking | — | 18% |
| DeepResearch Bench | 49.6% | — |
| BALROG | 32.8% | — |
| METR Time Horizons | 69.6% | — |
| Vending-Bench 2 | — | 4,663 |
Reasoning Grok 4.20 (Non-Reasoning) leads
GPT-5: 38.3 (#55), Grok 4.20 (Non-Reasoning): 52.2 (#30)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 9.9% | 65.1% |
| Kagi LLM Benchmark | 72.7% | 75% |
| ARC-AGI-1 | 65.7% | 89.5% |
| Chess Puzzles | 37% | 24% |
| LMArena Hard Prompts | 1415 | 1451 |
| DTBench | 90.7% | 90.1% |
| LMCA | 40% | 38.7% |
| Epoch Capabilities Index | 150 | 152.01 |
| ForecastBench | 61.4 | 61.4 |
| SimpleBench | 56.7% | — |
| NYT Connections (extended) | — | 85.4% |
| CritPt | 12.6% | — |
| EnigmaEval | 10.5% | — |
| Thematic Generalization | — | 63.8% |
| EBR-Bench | 12.7% | — |
| Mystery Game Puzzles | 23% | — |
Math GPT-5 leads
GPT-5: 54.6 (#36), Grok 4.20 (Non-Reasoning): 48.0 (#54)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| FrontierMath (Tiers 1-3) | 55.4% | 44.9% |
| FrontierMath Tier 4 | 22% | 17.1% |
| OTIS Mock AIME 2024-2025 | 91.4% | 92.2% |
| ProofBench | 18% | 14% |
| LMArena Math | 1406 | 1455 |
| Omni-MATH | 64.7% | — |
| MATH Level 5 | 98.1% | — |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GPT-5 leads
GPT-5: 56.5 (#37), Grok 4.20 (Non-Reasoning): 52.8 (#54)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 86.2% | 89.3% |
| SimpleQA Verified | 50.1% | 30.2% |
| LMArena Expert | 1420 | 1440 |
| Humanity's Last Exam | 25.3% | — |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.3% | — |
| Vectara Hallucination Rate | 14.7% | — |
| GPQA (HELM) | 79.2% | — |
Multimodal GPT-5 leads
GPT-5: 46.1 (#12), Grok 4.20 (Non-Reasoning): 33.2 (#88)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Vision | 1232 | 1263 |
| GeoBench | 81% | — |
| VPCT | 66% | — |
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1416 |
Multilingual Grok 4.20 (Non-Reasoning) leads
GPT-5: 51.5 (#105), Grok 4.20 (Non-Reasoning): 54.7 (#38)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1397 | 1440 |
| LMArena Chinese | 1423 | 1481 |
| LMArena French | 1409 | 1477 |
| LMArena German | 1415 | 1463 |
| LMArena Japanese | 1411 | 1450 |
| LMArena Korean | 1361 | 1417 |
| LMArena Russian | 1406 | 1457 |
| LMArena Spanish | 1398 | 1442 |
Instruction Following Grok 4.20 (Non-Reasoning) leads
GPT-5: 74.0 (#107), Grok 4.20 (Non-Reasoning): 75.0 (#79)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1388 | 1420 |
| IFEval | 87.5% | — |
Long Context GPT-5 leads
GPT-5: 69.5 (#2), Grok 4.20 (Non-Reasoning): 45.4 (#33)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Longer Query | 1399 | 1436 |
| Fiction.LiveBench | 97.2% | — |
| CL-bench | — | 22.2% |
| CL-bench Life | — | 11.9% |
Writing & Preference Grok 4.20 (Non-Reasoning) leads
GPT-5: 63.5 (#63), Grok 4.20 (Non-Reasoning): 65.9 (#42)
| Benchmark | GPT-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1406 | 1451 |
| LMArena Creative Writing | 1365 | 1439 |
| EQ-Bench Creative Writing | 1627 | 1574 |
| LMArena Multi-Turn | 1426 | 1456 |
| Short-Story Creative Writing | 86% | — |
| WildBench | 85.7% | — |
Frequently asked questions
Is GPT-5 better than Grok 4.20 (Non-Reasoning)?
GPT-5 is the stronger model overall, scoring 50.9 to 48.5 on the Noometry Index. Grok 4.20 (Non-Reasoning) costs 2.2× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.
Which is cheaper, GPT-5 or Grok 4.20 (Non-Reasoning)?
Grok 4.20 (Non-Reasoning) is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; GPT-5 lists at $1.25 and $10.
Is GPT-5 or Grok 4.20 (Non-Reasoning) better for coding?
GPT-5 scores higher on coding benchmarks: 50.7 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
Grok 4.20 (Non-Reasoning) does, with 1M tokens against 400K.
How many benchmarks do GPT-5 and Grok 4.20 (Non-Reasoning) share?
38 benchmarks have published results for both models. GPT-5 has 69 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.