Model comparison
GPT-5 Mini vs Grok 4.6
Grok 4.6 is the stronger model overall, scoring 56.9 to 41.8 on the Noometry Index. GPT-5 Mini costs 4.4× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GPT-5 Mini scores higher in 1 category and Grok 4.6 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 23.9.
- The biggest single-benchmark swing is ARC-AGI-2: 4.4% for GPT-5 Mini and 67.1% for Grok 4.6.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 400K.
Side by side
| GPT-5 Mini | Grok 4.6 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 41.8 | 56.9 |
| Released | 2025-08-07 | 2026-08-12 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 500K |
| Max output | 128K | 500K |
| Input $ / M tokens | $0.25 | $2 |
| Output $ / M tokens | $2 | $6 |
| Results tracked | 60 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
GPT-5 Mini: 40.1 (#146), Grok 4.6: 58.5 (#16)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| SciCode | 39.2% | 56.5% |
| WeirdML | 52.7% | 67.3% |
| LMArena Coding | 1406 | 1465 |
| ALE-Bench | 799.77 | 1,508 |
| SWE-bench Verified | 64.7% | — |
| DeepSWE | — | 67.5% |
| FrontierCode | — | 48% |
| SWE-bench Verified (bash only) | 59.8% | — |
| CursorBench | — | 41.4% |
| LMArena WebDev | — | 1617 |
| SWE-bench Multilingual | 39.7% | — |
| FrontierSWE | — | 25.3% |
| AlgoTune | 1.38 | — |
Agentic & Tool Use Grok 4.6 leads
GPT-5 Mini: 31.1 (#70), Grok 4.6: 39.4 (#27)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| Vending-Bench 2 | -31.18 | 9,047 |
| Terminal-Bench | 34.8% | — |
| APEX-Agents | — | 65.3% |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| GDP.pdf | — | 17.2% |
Reasoning Grok 4.6 leads
GPT-5 Mini: 23.9 (#168), Grok 4.6: 61.4 (#20)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| ARC-AGI-2 | 4.4% | 67.1% |
| ARC-AGI-1 | 54.3% | 87.5% |
| CritPt | 0% | 19.7% |
| Chess Puzzles | 30% | 40% |
| LMArena Hard Prompts | 1380 | 1447 |
| Mystery Game Puzzles | 10% | 34% |
| DTBench | 80.5% | 97.3% |
| LMCA | 34.2% | 48.5% |
| Epoch Capabilities Index | 145.52 | 156.44 |
| SimpleBench | — | 75.9% |
| Kagi LLM Benchmark | 70.3% | — |
| NYT Connections (extended) | — | 80% |
| EnigmaEval | 8.2% | — |
| EBR-Bench | — | 30.5% |
| ForecastBench | 61 | — |
Math Grok 4.6 leads
GPT-5 Mini: 46.7 (#69), Grok 4.6: 67.0 (#24)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.7% | 66% |
| FrontierMath Tier 4 | 12.2% | 31.7% |
| OTIS Mock AIME 2024-2025 | 86.7% | 99.2% |
| ProofBench | 9% | 51% |
| LMArena Math | 1378 | 1423 |
| Omni-MATH | 72.2% | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge Grok 4.6 leads
GPT-5 Mini: 45.6 (#86), Grok 4.6: 63.3 (#20)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 75% | 94% |
| SimpleQA Verified | 21.6% | 49.3% |
| LMArena Expert | 1379 | 1467 |
| Humanity's Last Exam | 19.4% | — |
| MMLU-Pro | 83.5% | — |
| Confabulations | 13.3% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |
Multimodal Grok 4.6 leads
GPT-5 Mini: 35.6 (#85), Grok 4.6: 43.6 (#23)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| LMArena Vision | 1202 | 1263 |
| VPCT | 40.2% | — |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| LMArena Document | — | 1452 |
Multilingual Grok 4.6 leads
GPT-5 Mini: 48.9 (#137), Grok 4.6: 53.0 (#74)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1363 | 1420 |
| LMArena Chinese | 1385 | 1480 |
| LMArena French | 1386 | 1461 |
| LMArena German | 1366 | 1431 |
| LMArena Japanese | 1341 | 1376 |
| LMArena Korean | 1308 | 1397 |
| LMArena Russian | 1362 | 1422 |
| LMArena Spanish | 1355 | 1404 |
Instruction Following Too close to call
GPT-5 Mini: 76.2 (#46), Grok 4.6: 75.4 (#63)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1357 | 1431 |
| IFEval | 92.7% | — |
Long Context Grok 4.6 leads
GPT-5 Mini: 41.9 (#132), Grok 4.6: 44.5 (#66)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1355 | 1454 |
| Fiction.LiveBench | 69.4% | — |
Writing & Preference Grok 4.6 leads
GPT-5 Mini: 55.2 (#148), Grok 4.6: 62.3 (#80)
| Benchmark | GPT-5 Mini | Grok 4.6 |
|---|---|---|
| LMArena Text | 1373 | 1428 |
| LMArena Creative Writing | 1325 | 1428 |
| LMArena Multi-Turn | 1363 | 1425 |
| Short-Story Creative Writing | 83.1% | — |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
Frequently asked questions
Is GPT-5 Mini better than Grok 4.6?
Grok 4.6 is the stronger model overall, scoring 56.9 to 41.8 on the Noometry Index. GPT-5 Mini costs 4.4× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Mini or Grok 4.6?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Grok 4.6 lists at $2 and $6.
Is GPT-5 Mini or Grok 4.6 better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 400K.
How many benchmarks do GPT-5 Mini and Grok 4.6 share?
36 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Grok 4.6 has 49.