Model comparison
Grok 4.20 (Non-Reasoning) vs Inkling
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 44.1 on the Noometry Index.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 7 categories and Inkling in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.20 (Non-Reasoning) leads 48.2 to 31.3.
- The biggest single-benchmark swing is ARC-AGI-2: 65.1% for Grok 4.20 (Non-Reasoning) and 36.5% for Inkling.
- Grok 4.20 (Non-Reasoning) is cheaper at $1.25 / $2.50 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.
Side by side
| Grok 4.20 (Non-Reasoning) | Inkling | |
|---|---|---|
| Provider | xAI | Thinking Machines Lab |
| Noometry Index | 48.6 | 44.1 |
| Released | 2026-02-17 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 1M | 66K |
| Max output | 30K | 66K |
| Input $ / M tokens | $1.25 | $1.87 |
| Output $ / M tokens | $2.50 | $4.68 |
| Results tracked | 46 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 42.1 (#112), Inkling: 34.5 (#234)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| LMArena WebDev | 1375 | 1413 |
| WeirdML | 52.3% | 32.3% |
| LMArena Coding | 1459 | 1464 |
| ALE-Bench | 1,150 | 946 |
| FrontierCode | — | 14% |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 34.4 (#46), Inkling: 29.6 (#85)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| τ²-bench Banking | 18% | 25% |
| Terminal-Bench | 57.3% | — |
| APEX-Agents | — | 33.8% |
| LMArena Search | 1189 | — |
| Vending-Bench 2 | 4,663 | — |
Reasoning Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 52.3 (#32), Inkling: 40.4 (#56)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| ARC-AGI-2 | 65.1% | 36.5% |
| ARC-AGI-1 | 89.5% | 79.5% |
| Chess Puzzles | 24% | 21% |
| LMArena Hard Prompts | 1451 | 1451 |
| DTBench | 90.1% | 87.5% |
| LMCA | 38.7% | 37.6% |
| Epoch Capabilities Index | 151.98 | 148.54 |
| SimpleBench | — | 50% |
| Kagi LLM Benchmark | 75% | — |
| NYT Connections (extended) | 85.4% | — |
| CritPt | — | 5.4% |
| Thematic Generalization | 63.8% | — |
| ForecastBench | 61.4 | — |
Math Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 48.2 (#65), Inkling: 31.3 (#225)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| FrontierMath (Tiers 1-3) | 44.9% | 33.3% |
| FrontierMath Tier 4 | 17.1% | 4.9% |
| OTIS Mock AIME 2024-2025 | 92.2% | 88.9% |
| ProofBench | 14% | 0% |
| LMArena Math | 1455 | 1479 |
Knowledge Inkling leads
Grok 4.20 (Non-Reasoning): 52.8 (#60), Inkling: 55.1 (#49)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| GPQA Diamond | 89.3% | 88.3% |
| SimpleQA Verified | 30.2% | 40.3% |
| LMArena Expert | 1439 | 1465 |
Multimodal Not comparable
Grok 4.20 (Non-Reasoning): 33.3 (#98), Inkling: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| LMArena Vision | 1263 | — |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1416 | — |
Multilingual Too close to call
Grok 4.20 (Non-Reasoning): 54.5 (#40), Inkling: 54.0 (#52)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| LMArena Non-English | 1441 | 1434 |
| LMArena Chinese | 1481 | 1490 |
| LMArena French | 1476 | 1458 |
| LMArena German | 1465 | 1446 |
| LMArena Japanese | 1449 | 1429 |
| LMArena Korean | 1417 | 1404 |
| LMArena Russian | 1458 | 1429 |
| LMArena Spanish | 1443 | 1448 |
Instruction Following Too close to call
Grok 4.20 (Non-Reasoning): 74.8 (#83), Inkling: 75.1 (#71)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| LMArena Instruction Following | 1420 | 1426 |
Long Context Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 45.5 (#34), Inkling: 43.8 (#86)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| LMArena Longer Query | 1437 | 1434 |
| CL-bench | 22.2% | — |
| CL-bench Life | 11.9% | — |
Writing & Preference Too close to call
Grok 4.20 (Non-Reasoning): 65.7 (#44), Inkling: 65.2 (#51)
| Benchmark | Grok 4.20 (Non-Reasoning) | Inkling |
|---|---|---|
| LMArena Text | 1451 | 1441 |
| LMArena Creative Writing | 1438 | 1387 |
| EQ-Bench Creative Writing | 1574 | 1611 |
| LMArena Multi-Turn | 1456 | 1436 |
| EQ-Bench 4 | — | 1226 |
Frequently asked questions
Is Grok 4.20 (Non-Reasoning) better than Inkling?
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 44.1 on the Noometry Index.
Which is cheaper, Grok 4.20 (Non-Reasoning) or Inkling?
Grok 4.20 (Non-Reasoning) is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; Inkling lists at $1.87 and $4.68.
Is Grok 4.20 (Non-Reasoning) or Inkling better for coding?
Grok 4.20 (Non-Reasoning) scores higher on coding benchmarks: 42.1 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Grok 4.20 (Non-Reasoning) does, with 1M tokens against 66K.
How many benchmarks do Grok 4.20 (Non-Reasoning) and Inkling share?
34 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and Inkling has 41.