Model comparison
Grok 4.3 vs Nemotron 3 Ultra
Grok 4.3 is the stronger model overall, scoring 43.8 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 1.7× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
NVIDIASummary
- They share 26 benchmarks with published results for both. Grok 4.3 scores higher in 4 categories and Nemotron 3 Ultra in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.3 leads 46.0 to 35.3.
- The biggest single-benchmark swing is Chess Puzzles: 25% for Grok 4.3 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 4.3.
- Grok 4.3 accepts more context: 1M tokens versus 262K.
- Nemotron 3 Ultra has downloadable open weights; the other is API-only.
Side by side
| Grok 4.3 | Nemotron 3 Ultra | |
|---|---|---|
| Provider | xAI | NVIDIA |
| Noometry Index | 43.8 | 42.5 |
| Released | 2026-04-17 | 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 | 40 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.3 leads
Grok 4.3: 41.6 (#121), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| SciCode | 47.3% | 40.3% |
| WeirdML | 49.9% | 43.5% |
| LMArena Coding | 1415 | 1468 |
| FrontierCode | — | 13.6% |
| LMArena WebDev | 1357 | — |
| ALE-Bench | 944.17 | — |
Agentic & Tool Use Grok 4.3 leads
Grok 4.3: 27.7 (#99), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | — | 22.7% |
| GDP.pdf | 8% | — |
| LMArena Search | 1165 | — |
| Vending-Bench 2 | 35.26 | — |
Reasoning Grok 4.3 leads
Grok 4.3: 35.9 (#68), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 8% | 3.1% |
| Chess Puzzles | 25% | 12% |
| LMArena Hard Prompts | 1396 | 1452 |
| DTBench | 90.7% | 90.1% |
| LMCA | 38.3% | 36.9% |
| Epoch Capabilities Index | 149.16 | 146.17 |
| NYT Connections (extended) | 55.2% | — |
| Mystery Game Puzzles | — | 20% |
| ForecastBench | 60.3 | — |
Math Grok 4.3 leads
Grok 4.3: 46.0 (#74), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 93.3% | 86.7% |
| ProofBench | 11% | 2% |
| LMArena Math | 1388 | 1457 |
| FrontierMath (Tiers 1-3) | 42.8% | — |
| FrontierMath Tier 4 | 14.6% | — |
Knowledge Too close to call
Grok 4.3: 52.5 (#62), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 88.8% | 85.4% |
| LMArena Expert | 1385 | 1472 |
| SimpleQA Verified | 33.2% | — |
Multimodal Not comparable
Grok 4.3: 31.6 (#104), Nemotron 3 Ultra: —
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Vision | 1229 | — |
| Blueprint-Bench 2 | 0% | — |
Multilingual Nemotron 3 Ultra leads
Grok 4.3: 50.5 (#120), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1385 | 1427 |
| LMArena Chinese | 1422 | 1497 |
| LMArena French | 1412 | 1472 |
| LMArena German | 1395 | 1471 |
| LMArena Korean | 1356 | 1386 |
| LMArena Russian | 1399 | 1417 |
| LMArena Spanish | 1398 | 1454 |
| LMArena Japanese | 1379 | — |
Instruction Following Nemotron 3 Ultra leads
Grok 4.3: 72.1 (#140), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1366 | 1418 |
Long Context Nemotron 3 Ultra leads
Grok 4.3: 42.5 (#123), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1393 | 1437 |
Writing & Preference Nemotron 3 Ultra leads
Grok 4.3: 58.5 (#118), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | Grok 4.3 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1397 | 1445 |
| LMArena Creative Writing | 1380 | 1398 |
| LMArena Multi-Turn | 1406 | 1411 |
| EQ-Bench Creative Writing | — | 1692 |
| EQ-Bench 4 | 1075 | — |
Frequently asked questions
Is Grok 4.3 better than Nemotron 3 Ultra?
Grok 4.3 is the stronger model overall, scoring 43.8 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 1.7× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Which is cheaper, Grok 4.3 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 4.3 lists at $1.25 and $2.50.
Is Grok 4.3 or Nemotron 3 Ultra better for coding?
Grok 4.3 scores higher on coding benchmarks: 41.6 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
Grok 4.3 does, with 1M tokens against 262K.
How many benchmarks do Grok 4.3 and Nemotron 3 Ultra share?
26 benchmarks have published results for both models. Grok 4.3 has 40 scored results on Noometry and Nemotron 3 Ultra has 30.