Model comparison
Grok 4.20 Multi-Agent vs Inkling
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 44.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
Summary
- They share 17 benchmarks with published results for both. Grok 4.20 Multi-Agent scores higher in 4 categories and Inkling in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.1 to 40.4.
- Grok 4.20 Multi-Agent is cheaper at $1.25 / $2.50 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Grok 4.20 Multi-Agent accepts more context: 1M tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.
Side by side
| Grok 4.20 Multi-Agent | Inkling | |
|---|---|---|
| Provider | xAI | Thinking Machines Lab |
| Noometry Index | 46.2 | 44.1 |
| Released | 2026-03-09 | 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 | 20 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 43.0 (#92), Inkling: 34.5 (#234)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Coding | 1457 | 1464 |
| FrontierCode | — | 14% |
| LMArena WebDev | — | 1413 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| WeirdML | — | 32.3% |
| ALE-Bench | — | 946 |
Agentic & Tool Use Not comparable
Grok 4.20 Multi-Agent: —, Inkling: 29.6 (#85)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |
| LMArena Search | 1204 | — |
Reasoning Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 43.9 (#48), Inkling: 40.4 (#56)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Hard Prompts | 1448 | 1451 |
| ARC-AGI-2 | — | 36.5% |
| SimpleBench | — | 50% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 79.5% |
| CritPt | — | 5.4% |
| Chess Puzzles | — | 21% |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |
| Epoch Capabilities Index | — | 148.54 |
Math Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 39.4 (#104), Inkling: 31.3 (#225)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Math | 1442 | 1479 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 0% |
Knowledge Inkling leads
Grok 4.20 Multi-Agent: 40.4 (#119), Inkling: 55.1 (#49)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Expert | 1445 | 1465 |
| GPQA Diamond | — | 88.3% |
| SimpleQA Verified | — | 40.3% |
Multimodal Not comparable
Grok 4.20 Multi-Agent: 40.5 (#48), Inkling: —
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Vision | 1259 | — |
Multilingual Too close to call
Grok 4.20 Multi-Agent: 54.4 (#43), Inkling: 54.0 (#52)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Non-English | 1440 | 1434 |
| LMArena Chinese | 1475 | 1490 |
| LMArena French | 1466 | 1458 |
| LMArena German | 1456 | 1446 |
| LMArena Japanese | 1405 | 1429 |
| LMArena Korean | 1416 | 1404 |
| LMArena Russian | 1457 | 1429 |
| LMArena Spanish | 1447 | 1448 |
Instruction Following Too close to call
Grok 4.20 Multi-Agent: 74.8 (#84), Inkling: 75.1 (#71)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Instruction Following | 1420 | 1426 |
Long Context Too close to call
Grok 4.20 Multi-Agent: 43.7 (#88), Inkling: 43.8 (#86)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Longer Query | 1431 | 1434 |
Writing & Preference Inkling leads
Grok 4.20 Multi-Agent: 64.0 (#59), Inkling: 65.2 (#51)
| Benchmark | Grok 4.20 Multi-Agent | Inkling |
|---|---|---|
| LMArena Text | 1450 | 1441 |
| LMArena Creative Writing | 1436 | 1387 |
| LMArena Multi-Turn | 1452 | 1436 |
| EQ-Bench Creative Writing | — | 1611 |
| EQ-Bench 4 | — | 1226 |
Frequently asked questions
Is Grok 4.20 Multi-Agent better than Inkling?
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 44.1 on the Noometry Index.
Which is cheaper, Grok 4.20 Multi-Agent or Inkling?
Grok 4.20 Multi-Agent 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 Multi-Agent or Inkling better for coding?
Grok 4.20 Multi-Agent scores higher on coding benchmarks: 43.0 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Grok 4.20 Multi-Agent does, with 1M tokens against 66K.
How many benchmarks do Grok 4.20 Multi-Agent and Inkling share?
17 benchmarks have published results for both models. Grok 4.20 Multi-Agent has 20 scored results on Noometry and Inkling has 41.