Model comparison
GPT-4 Turbo vs Grok 3
Grok 3 is the stronger model overall, scoring 39.9 to 30.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. GPT-4 Turbo scores higher in 1 category and Grok 3 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 3 leads 38.0 to 9.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 55.6% for Grok 3.
Side by side
| GPT-4 Turbo | Grok 3 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 30.5 | 39.9 |
| Released | 2023-11-06 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $10 | — |
| Output $ / M tokens | $30 | — |
| Results tracked | 36 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
GPT-4 Turbo: 33.8 (#249), Grok 3: 41.9 (#115)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| WeirdML | 18% | 37.2% |
| LMArena Coding | 1268 | 1432 |
| Aider Polyglot | — | 53.3% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, Grok 3: 30.5 (#76)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| BALROG | — | 29.5% |
| METR Time Horizons | 36.7% | — |
Reasoning GPT-4 Turbo leads
GPT-4 Turbo: 15.3 (#317), Grok 3: 13.7 (#333)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| SimpleBench | 25.1% | 36.1% |
| LMArena Hard Prompts | 1251 | 1434 |
| Epoch Capabilities Index | 127.25 | 138.33 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 61.3% |
| ARC-AGI-1 | — | 5.5% |
| Chess Puzzles | 6% | — |
| DTBench | 61.6% | — |
| LMCA | 9.8% | — |
| ForecastBench | 59.4 | — |
Math Grok 3 leads
GPT-4 Turbo: 9.0 (#322), Grok 3: 38.0 (#145)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.7% | 55.6% |
| LMArena Math | 1272 | 1391 |
| MATH Level 5 | 46.7% | 88.7% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| Omni-MATH | — | 46.4% |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Grok 3 leads
GPT-4 Turbo: 24.3 (#268), Grok 3: 46.2 (#82)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| GPQA Diamond | 46.6% | 75.8% |
| Confabulations | 28.4% | 14.2% |
| LMArena Expert | 1223 | 1421 |
| MMLU-Pro | — | 78.8% |
| Vectara Hallucination Rate | — | 5.8% |
| GPQA (HELM) | — | 65% |
| MMLU | 81.3% | — |
Multimodal Not comparable
GPT-4 Turbo: 30.6 (#110), Grok 3: —
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Vision | 1090 | — |
Multilingual Grok 3 leads
GPT-4 Turbo: 40.5 (#216), Grok 3: 52.3 (#87)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Non-English | 1245 | 1410 |
| LMArena Chinese | 1242 | 1448 |
| LMArena French | 1276 | 1460 |
| LMArena German | 1259 | 1431 |
| LMArena Japanese | 1194 | 1387 |
| LMArena Korean | 1187 | 1373 |
| LMArena Russian | 1259 | 1416 |
| LMArena Spanish | 1260 | 1417 |
Instruction Following Grok 3 leads
GPT-4 Turbo: 65.8 (#216), Grok 3: 75.0 (#73)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1249 | 1409 |
| IFEval | — | 88.4% |
Long Context Too close to call
GPT-4 Turbo: 38.0 (#206), Grok 3: 38.7 (#192)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1254 | 1439 |
| Fiction.LiveBench | — | 58.3% |
Writing & Preference Grok 3 leads
GPT-4 Turbo: 47.7 (#206), Grok 3: 55.8 (#141)
| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Text | 1272 | 1426 |
| LMArena Creative Writing | 1269 | 1414 |
| LMArena Multi-Turn | 1267 | 1425 |
| Short-Story Creative Writing | — | 76.4% |
| EQ-Bench Creative Writing | — | 1186 |
| WildBench | — | 84.9% |
Frequently asked questions
Is GPT-4 Turbo better than Grok 3?
Grok 3 is the stronger model overall, scoring 39.9 to 30.5 on the Noometry Index.
Is GPT-4 Turbo or Grok 3 better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 33.8 in the Noometry coding category.
How many benchmarks do GPT-4 Turbo and Grok 3 share?
24 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Grok 3 has 40.