Model comparison
Grok 4.20 (Non-Reasoning) vs Nemotron 3 Ultra
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 1.7× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
NVIDIASummary
- They share 25 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 8 categories and Nemotron 3 Ultra in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 31.0.
- The biggest single-benchmark swing is Chess Puzzles: 24% for Grok 4.20 (Non-Reasoning) and 12% for Nemotron 3 Ultra.
- Nemotron 3 Ultra is cheaper at $0.50 / $2.20 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 (Non-Reasoning).
- Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 262K.
- Nemotron 3 Ultra has downloadable open weights; the other is API-only.
Side by side
| Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra | |
|---|---|---|
| Provider | xAI | NVIDIA |
| Noometry Index | 48.6 | 42.5 |
| Released | 2026-02-17 | 2026-06-04 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 30K | 128K |
| Input $ / M tokens | $1.25 | $0.50 |
| Output $ / M tokens | $2.50 | $2.20 |
| Results tracked | 46 | 30 |
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), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| WeirdML | 52.3% | 43.5% |
| LMArena Coding | 1459 | 1468 |
| FrontierCode | — | 13.6% |
| LMArena WebDev | 1375 | — |
| SciCode | — | 40.3% |
| ALE-Bench | 1,150 | — |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 34.4 (#46), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| Terminal-Bench | 57.3% | — |
| APEX-Agents | — | 22.7% |
| τ²-bench Banking | 18% | — |
| LMArena Search | 1189 | — |
| Vending-Bench 2 | 4,663 | — |
Reasoning Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 52.3 (#32), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| Chess Puzzles | 24% | 12% |
| LMArena Hard Prompts | 1451 | 1452 |
| DTBench | 90.1% | 90.1% |
| LMCA | 38.7% | 36.9% |
| Epoch Capabilities Index | 151.98 | 146.17 |
| ARC-AGI-2 | 65.1% | — |
| Kagi LLM Benchmark | 75% | — |
| NYT Connections (extended) | 85.4% | — |
| ARC-AGI-1 | 89.5% | — |
| CritPt | — | 3.1% |
| Thematic Generalization | 63.8% | — |
| Mystery Game Puzzles | — | 20% |
| ForecastBench | 61.4 | — |
Math Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 48.2 (#65), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 86.7% |
| ProofBench | 14% | 2% |
| LMArena Math | 1455 | 1457 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 17.1% | — |
Knowledge Too close to call
Grok 4.20 (Non-Reasoning): 52.8 (#60), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 89.3% | 85.4% |
| LMArena Expert | 1439 | 1472 |
| SimpleQA Verified | 30.2% | — |
Multimodal Not comparable
Grok 4.20 (Non-Reasoning): 33.3 (#98), Nemotron 3 Ultra: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| LMArena Vision | 1263 | — |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1416 | — |
Multilingual Too close to call
Grok 4.20 (Non-Reasoning): 54.5 (#40), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1441 | 1427 |
| LMArena Chinese | 1481 | 1497 |
| LMArena French | 1476 | 1472 |
| LMArena German | 1465 | 1471 |
| LMArena Korean | 1417 | 1386 |
| LMArena Russian | 1458 | 1417 |
| LMArena Spanish | 1443 | 1454 |
| LMArena Japanese | 1449 | — |
Instruction Following Too close to call
Grok 4.20 (Non-Reasoning): 74.8 (#83), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1420 | 1418 |
Long Context Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 45.5 (#34), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1437 | 1437 |
| CL-bench | 22.2% | — |
| CL-bench Life | 11.9% | — |
Writing & Preference Too close to call
Grok 4.20 (Non-Reasoning): 65.7 (#44), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | Grok 4.20 (Non-Reasoning) | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1451 | 1445 |
| LMArena Creative Writing | 1438 | 1398 |
| EQ-Bench Creative Writing | 1574 | 1692 |
| LMArena Multi-Turn | 1456 | 1411 |
Frequently asked questions
Is Grok 4.20 (Non-Reasoning) better than Nemotron 3 Ultra?
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 1.7× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.
Which is cheaper, Grok 4.20 (Non-Reasoning) or Nemotron 3 Ultra?
Nemotron 3 Ultra is cheaper. It lists at $0.50 per million input tokens and $2.20 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.
Is Grok 4.20 (Non-Reasoning) or Nemotron 3 Ultra better for coding?
Grok 4.20 (Non-Reasoning) scores higher on coding benchmarks: 42.1 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
Grok 4.20 (Non-Reasoning) does, with 1M tokens against 262K.
How many benchmarks do Grok 4.20 (Non-Reasoning) and Nemotron 3 Ultra share?
25 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and Nemotron 3 Ultra has 30.