Model comparison
Grok 4.6 vs Inkling
Grok 4.6 is the stronger model overall, scoring 56.9 to 44.1 on the Noometry Index.
Last verified . 38 shared benchmarks.
Summary
- They share 38 benchmarks with published results for both. Grok 4.6 scores higher in 7 categories and Inkling in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.6 leads 67.0 to 31.3.
- The biggest single-benchmark swing is ProofBench: 51% for Grok 4.6 and 0% for Inkling.
- Inkling is cheaper at $1.87 / $4.68 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.
Side by side
| Grok 4.6 | Inkling | |
|---|---|---|
| Provider | xAI | Thinking Machines Lab |
| Noometry Index | 56.9 | 44.1 |
| Released | 2026-08-12 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 500K | 66K |
| Max output | 500K | 66K |
| Input $ / M tokens | $2 | $1.87 |
| Output $ / M tokens | $6 | $4.68 |
| Results tracked | 49 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
Grok 4.6: 58.5 (#16), Inkling: 34.5 (#234)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| FrontierCode | 48% | 14% |
| LMArena WebDev | 1617 | 1413 |
| FrontierSWE | 25.3% | 4.1% |
| SciCode | 56.5% | 47% |
| WeirdML | 67.3% | 32.3% |
| LMArena Coding | 1465 | 1464 |
| ALE-Bench | 1,508 | 946 |
| DeepSWE | 67.5% | — |
| CursorBench | 41.4% | — |
Agentic & Tool Use Grok 4.6 leads
Grok 4.6: 39.4 (#27), Inkling: 29.6 (#85)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| APEX-Agents | 65.3% | 33.8% |
| τ²-bench Banking | — | 25% |
| GDP.pdf | 17.2% | — |
| Vending-Bench 2 | 9,047 | — |
Reasoning Grok 4.6 leads
Grok 4.6: 61.4 (#20), Inkling: 40.4 (#56)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| ARC-AGI-2 | 67.1% | 36.5% |
| SimpleBench | 75.9% | 50% |
| ARC-AGI-1 | 87.5% | 79.5% |
| CritPt | 19.7% | 5.4% |
| Chess Puzzles | 40% | 21% |
| LMArena Hard Prompts | 1447 | 1451 |
| DTBench | 97.3% | 87.5% |
| LMCA | 48.5% | 37.6% |
| Epoch Capabilities Index | 156.44 | 148.54 |
| NYT Connections (extended) | 80% | — |
| EBR-Bench | 30.5% | — |
| Mystery Game Puzzles | 34% | — |
Math Grok 4.6 leads
Grok 4.6: 67.0 (#24), Inkling: 31.3 (#225)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| FrontierMath (Tiers 1-3) | 66% | 33.3% |
| FrontierMath Tier 4 | 31.7% | 4.9% |
| OTIS Mock AIME 2024-2025 | 99.2% | 88.9% |
| ProofBench | 51% | 0% |
| LMArena Math | 1423 | 1479 |
Knowledge Grok 4.6 leads
Grok 4.6: 63.3 (#20), Inkling: 55.1 (#49)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| GPQA Diamond | 94% | 88.3% |
| SimpleQA Verified | 49.3% | 40.3% |
| LMArena Expert | 1467 | 1465 |
Multimodal Not comparable
Grok 4.6: 43.6 (#23), Inkling: —
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| LMArena Vision | 1263 | — |
| Blueprint-Bench 2 | 33.2% | — |
| Furniture Assembly | 40% | — |
| LMArena Document | 1452 | — |
Multilingual Inkling leads
Grok 4.6: 53.0 (#74), Inkling: 54.0 (#52)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| LMArena Non-English | 1420 | 1434 |
| LMArena Chinese | 1480 | 1490 |
| LMArena French | 1461 | 1458 |
| LMArena German | 1431 | 1446 |
| LMArena Japanese | 1376 | 1429 |
| LMArena Korean | 1397 | 1404 |
| LMArena Russian | 1422 | 1429 |
| LMArena Spanish | 1404 | 1448 |
Instruction Following Too close to call
Grok 4.6: 75.4 (#63), Inkling: 75.1 (#71)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| LMArena Instruction Following | 1431 | 1426 |
Long Context Too close to call
Grok 4.6: 44.5 (#66), Inkling: 43.8 (#86)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| LMArena Longer Query | 1454 | 1434 |
Writing & Preference Inkling leads
Grok 4.6: 62.3 (#80), Inkling: 65.2 (#51)
| Benchmark | Grok 4.6 | Inkling |
|---|---|---|
| LMArena Text | 1428 | 1441 |
| LMArena Creative Writing | 1428 | 1387 |
| LMArena Multi-Turn | 1425 | 1436 |
| EQ-Bench Creative Writing | — | 1611 |
| EQ-Bench 4 | — | 1226 |
Frequently asked questions
Is Grok 4.6 better than Inkling?
Grok 4.6 is the stronger model overall, scoring 56.9 to 44.1 on the Noometry Index.
Which is cheaper, Grok 4.6 or Inkling?
Inkling is cheaper. It lists at $1.87 per million input tokens and $4.68 per million output tokens; Grok 4.6 lists at $2 and $6.
Is Grok 4.6 or Inkling better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 66K.
How many benchmarks do Grok 4.6 and Inkling share?
38 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and Inkling has 41.