Model comparison
GPT-4 Turbo vs Grok 4.6
Grok 4.6 is the stronger model overall, scoring 56.9 to 30.5 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and Grok 4.6 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.6 leads 67.0 to 9.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 99.2% for Grok 4.6.
- Grok 4.6 is cheaper at $2 / $6 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- Grok 4.6 accepts more context: 500K tokens versus 128K.
Side by side
| GPT-4 Turbo | Grok 4.6 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 30.5 | 56.9 |
| Released | 2023-11-06 | 2026-08-12 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 500K |
| Max output | 4K | 500K |
| Input $ / M tokens | $10 | $2 |
| Output $ / M tokens | $30 | $6 |
| Results tracked | 36 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
GPT-4 Turbo: 33.8 (#249), Grok 4.6: 58.5 (#16)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| WeirdML | 18% | 67.3% |
| LMArena Coding | 1268 | 1465 |
| DeepSWE | — | 67.5% |
| FrontierCode | — | 48% |
| CursorBench | — | 41.4% |
| LMArena WebDev | — | 1617 |
| FrontierSWE | — | 25.3% |
| SciCode | — | 56.5% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| ALE-Bench | — | 1,508 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, Grok 4.6: 39.4 (#27)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| APEX-Agents | — | 65.3% |
| GDP.pdf | — | 17.2% |
| METR Time Horizons | 36.7% | — |
| Vending-Bench 2 | — | 9,047 |
Reasoning Grok 4.6 leads
GPT-4 Turbo: 15.3 (#317), Grok 4.6: 61.4 (#20)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| SimpleBench | 25.1% | 75.9% |
| Chess Puzzles | 6% | 40% |
| LMArena Hard Prompts | 1251 | 1447 |
| DTBench | 61.6% | 97.3% |
| LMCA | 9.8% | 48.5% |
| Epoch Capabilities Index | 127.25 | 156.44 |
| ARC-AGI-2 | — | 67.1% |
| NYT Connections (extended) | — | 80% |
| ARC-AGI-1 | — | 87.5% |
| CritPt | — | 19.7% |
| EBR-Bench | — | 30.5% |
| Mystery Game Puzzles | — | 34% |
| ForecastBench | 59.4 | — |
Math Grok 4.6 leads
GPT-4 Turbo: 9.0 (#322), Grok 4.6: 67.0 (#24)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.7% | 66% |
| OTIS Mock AIME 2024-2025 | 6.7% | 99.2% |
| LMArena Math | 1272 | 1423 |
| FrontierMath Tier 4 | — | 31.7% |
| ProofBench | — | 51% |
| MATH Level 5 | 46.7% | — |
Knowledge Grok 4.6 leads
GPT-4 Turbo: 24.3 (#268), Grok 4.6: 63.3 (#20)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 46.6% | 94% |
| LMArena Expert | 1223 | 1467 |
| SimpleQA Verified | — | 49.3% |
| Confabulations | 28.4% | — |
| MMLU | 81.3% | — |
Multimodal Grok 4.6 leads
GPT-4 Turbo: 30.6 (#110), Grok 4.6: 43.6 (#23)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| LMArena Vision | 1090 | 1263 |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| LMArena Document | — | 1452 |
Multilingual Grok 4.6 leads
GPT-4 Turbo: 40.5 (#216), Grok 4.6: 53.0 (#74)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1245 | 1420 |
| LMArena Chinese | 1242 | 1480 |
| LMArena French | 1276 | 1461 |
| LMArena German | 1259 | 1431 |
| LMArena Japanese | 1194 | 1376 |
| LMArena Korean | 1187 | 1397 |
| LMArena Russian | 1259 | 1422 |
| LMArena Spanish | 1260 | 1404 |
Instruction Following Grok 4.6 leads
GPT-4 Turbo: 65.8 (#216), Grok 4.6: 75.4 (#63)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1249 | 1431 |
Long Context Grok 4.6 leads
GPT-4 Turbo: 38.0 (#206), Grok 4.6: 44.5 (#66)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1254 | 1454 |
Writing & Preference Grok 4.6 leads
GPT-4 Turbo: 47.7 (#206), Grok 4.6: 62.3 (#80)
| Benchmark | GPT-4 Turbo | Grok 4.6 |
|---|---|---|
| LMArena Text | 1272 | 1428 |
| LMArena Creative Writing | 1269 | 1428 |
| LMArena Multi-Turn | 1267 | 1425 |
Frequently asked questions
Is GPT-4 Turbo better than Grok 4.6?
Grok 4.6 is the stronger model overall, scoring 56.9 to 30.5 on the Noometry Index.
Which is cheaper, GPT-4 Turbo or Grok 4.6?
Grok 4.6 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is GPT-4 Turbo or Grok 4.6 better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 128K.
How many benchmarks do GPT-4 Turbo and Grok 4.6 share?
27 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Grok 4.6 has 49.