Grok 4.20 vs Nemotron 3 Ultra
Grok 4.20 leads Nemotron 3 Ultra on coding, 33.3 to 30.0 on the Vibe Coding Index, at about the same output price ($2.50 vs $2.40 per million). Updated October 22, 2026; 33 benchmarks have a result for both.
Grok 4.20 or Nemotron 3 Ultra: which to pick
Pick Grok 4.20 for coding: it scores 33.3 on the Vibe Coding Index against 30.0 for Nemotron 3 Ultra, though on 2 benchmarks against 3.
Pick Nemotron 3 Ultra for output tokens 1.0x cheaper ($2.40 vs $2.50 per million) and work measured by LiveCodeBench (Vals AI) or Vibe Code Bench (Vals AI), where it leads.
- Shared results
- 33
- Grok 4.20 only
- 22
- Nemotron 3 Ultra only
- 38
- Like-for-like categories
- 7 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
Grok 4.20
Grok 4.20 scores 33.3 against 30.0.
2 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
Grok 4.20
Grok 4.20 leads, 25.7 against 22.9.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Nemotron 3 Ultra has a AA Agentic Index result, 20.1.
Directional only
Reasoning
GPQA Diamond
Grok 4.20
Grok 4.20 leads, 88.6% against 86.1%.
Same board, both measured
Math
AIME
No call
Only Grok 4.20 has a AIME result, 96.5%.
Directional only
Knowledge
MMLU Pro
Grok 4.20
Grok 4.20 leads, 86.3% against 85.8%.
Same board, both measured
Web design
Design Arena: websites
Grok 4.20
Grok 4.20 leads, 1229 against 1145.
Same board, both measured
Long documents
Listed context window
Grok 4.20
Grok 4.20 takes 2M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Nemotron 3 Ultra
Grok 4.20 $0.0025, Nemotron 3 Ultra $0.0018 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Grok 4.20 has the number, $0.72.
Directional only
At a glance
- Coding index
- Grok 4.20 33.3 from 2 benchmarks, Nemotron 3 Ultra 30.0 from 3 benchmarks.
- Widest category
- Agentic work, 10.1 points apart on average over 4 shared benchmarks.
- Widest single gap
- Excel Modeling Benchmark (Vals AI): Grok 4.20 11.7% against Nemotron 3 Ultra 31.7%, 20.0 points apart on a 0 to 100 scale.
- Output price
- Grok 4.20 $2.50 against Nemotron 3 Ultra $2.40 per million output tokens, 1.0x apart.
- Context window
- Grok 4.20 takes 2M tokens, Nemotron 3 Ultra 262K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Grok 4.20 | Bars | Nemotron 3 Ultra | Ties |
|---|---|---|---|---|
| Coding | 3 | 4 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 1 | 3 | ||
| Reasoning | 3 | 0 | ||
| Knowledge | 2 | 3 | ||
| Web design | 8 | 0 | ||
| General | 1 | 2 | 1 tie | |
| All shared boards | 20 | 12 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index34
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| AA Coding Index | Not measured | 49.3 | Only Nemotron 3 Ultra measured |
| LiveCodeBench (Vals AI) | 84.3% | 86.0% | Nemotron 3 Ultra |
| SWE-bench Verified (Vals AI) | 72.2% | 69.0% | Grok 4.20 |
| Vibe Code Bench (Vals AI) | 4.1% | 7.6% | Nemotron 3 Ultra |
| Code migration (Vals AI) | 0.3% | 4.9% | Nemotron 3 Ultra |
| ProgramBench (Vals AI) | Not measured | 0.0% | Only Nemotron 3 Ultra measured |
| Terminal-Bench 2.1 (Vals AI) | 44.2% | 50.9% | Nemotron 3 Ultra |
| SciCode | Not measured | 40.3% | Only Nemotron 3 Ultra measured |
| Terminal-Bench Hard (Artificial Analysis run) | 37.9% | 36.4% | Grok 4.20 |
| Terminal-Bench | 57.3% | Not measured | Only Grok 4.20 measured |
| FrontierCode | Not measured | 13.6% | Only Nemotron 3 Ultra measured |
| WeirdML | 52.3% | 43.5% | Grok 4.20 |
| LMArena Text: coding | Not measured | 1468 | Only Nemotron 3 Ultra measured |
| LMArena Text: software and IT | Not measured | 1468 | Only Nemotron 3 Ultra measured |
| Neither model has a result on the other 29 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| AA Intelligence Index | 25.7 | 22.9 | Grok 4.20 |
| Epoch Capabilities Index | 152.1 | 146.0 | Grok 4.20 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index13
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 20.1 | Only Nemotron 3 Ultra measured |
| Design Arena: agents building web apps | 1144 | Not measured | Only Grok 4.20 measured |
| Design Arena: agents building full-stack apps | 1054 | Not measured | Only Grok 4.20 measured |
| Design Arena: agents building mobile apps | 1090 | Not measured | Only Grok 4.20 measured |
| Design Arena: agents building Android apps | 1045 | Not measured | Only Grok 4.20 measured |
| Design Arena: agents building Godot games | 1089 | Not measured | Only Grok 4.20 measured |
| Design Arena: HTML slides | 1180 | Not measured | Only Grok 4.20 measured |
| Finance Agent (Vals AI) | 28.5% | 37.7% | Nemotron 3 Ultra |
| Tax Agent Bench (Vals AI) | 7.0% | Not measured | Only Grok 4.20 measured |
| Legal Research Bench (Vals AI) | 13.9% | 15.4% | Nemotron 3 Ultra |
| Excel Modeling Benchmark (Vals AI) | 11.7% | 31.7% | Nemotron 3 Ultra |
| τ²-Bench Telecom | 93.0% | 83.3% | Grok 4.20 |
| APEX-Agents | Not measured | 22.7% | Only Nemotron 3 Ultra measured |
| Vending-Bench 2 | $4,663 | Not measured | Only Grok 4.20 measured |
| MCP Atlas | Not measured | 63.1% | Only Nemotron 3 Ultra measured |
| Neither model has a result on the other 27 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| GPQA Diamond | 88.6% | 86.1% | Grok 4.20 |
| MMMU Pro (Vals AI) | 83.5% | Not measured | Only Grok 4.20 measured |
| GPQA Diamond (Artificial Analysis run) | 91.1% | 86.7% | Grok 4.20 |
| Humanity's Last Exam (Artificial Analysis run) | 34.5% | 28.4% | Grok 4.20 |
| GPQA Diamond (Epoch AI run) | Not measured | 85.4% | Only Nemotron 3 Ultra measured |
| ARC-AGI | 89.5% | Not measured | Only Grok 4.20 measured |
| ARC-AGI-2 | 65.1% | Not measured | Only Grok 4.20 measured |
| Mystery game puzzles | Not measured | 20.0% | Only Nemotron 3 Ultra measured |
| LMArena Text: hard prompts | Not measured | 1452 | Only Nemotron 3 Ultra measured |
| LMArena Text: expert | Not measured | 1472 | Only Nemotron 3 Ultra measured |
| LMArena Text: hard prompts in English | Not measured | 1453 | Only Nemotron 3 Ultra measured |
| Kagi LLM Benchmark | 75.0% | Not measured | Only Grok 4.20 measured |
| Neither model has a result on the other 11 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| AIME | 96.5% | Not measured | Only Grok 4.20 measured |
| OTIS Mock AIME | Not measured | 86.7% | Only Nemotron 3 Ultra measured |
| ProofBench | Not measured | 2.0% | Only Nemotron 3 Ultra measured |
| LMArena Text: math | Not measured | 1457 | Only Nemotron 3 Ultra measured |
| LMArena Text: mathematical work | Not measured | 1461 | Only Nemotron 3 Ultra measured |
| Neither model has a result on the other 9 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index23
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| MMLU Pro | 86.3% | 85.8% | Grok 4.20 |
| LegalBench (Vals AI) | 77.7% | 82.1% | Nemotron 3 Ultra |
| CaseLaw (Vals AI) | 54.4% | Not measured | Only Grok 4.20 measured |
| CorpFin (Vals AI) | 63.7% | 65.5% | Nemotron 3 Ultra |
| TaxEval (Vals AI) | 74.1% | 73.1% | Grok 4.20 |
| MortgageTax (Vals AI) | 45.4% | Not measured | Only Grok 4.20 measured |
| MedQA (Vals AI) | 94.5% | Not measured | Only Grok 4.20 measured |
| MedCode (Vals AI) | 32.2% | 38.6% | Nemotron 3 Ultra |
| LMArena Text: science | Not measured | 1465 | Only Nemotron 3 Ultra measured |
| LMArena Text: medicine and healthcare | Not measured | 1456 | Only Nemotron 3 Ultra measured |
| LMArena Text: legal and government | Not measured | 1452 | Only Nemotron 3 Ultra measured |
| LMArena Text: business and finance | Not measured | 1434 | Only Nemotron 3 Ultra measured |
| Neither model has a result on the other 8 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index80
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| Design Arena: websites | 1229 | 1145 | Grok 4.20 |
| Design Arena: UI components | 1197 | 1135 | Grok 4.20 |
| Design Arena: game dev | 1193 | 1138 | Grok 4.20 |
| Design Arena: data viz | 1217 | 1141 | Grok 4.20 |
| Design Arena: 3D | 1193 | 1131 | Grok 4.20 |
| Design Arena: all code categories | 1221 | 1146 | Grok 4.20 |
| Design Arena: SVG | 1165 | 1069 | Grok 4.20 |
| Design Arena: ASCII art | 1175 | 1100 | Grok 4.20 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index12
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 38.2% | Not measured | Only Grok 4.20 measured |
| MedScribe (Vals AI) | 63.4% | Not measured | Only Grok 4.20 measured |
| Vals Multimodal Index | 39.1% | Not measured | Only Grok 4.20 measured |
| IFBench | 81.2% | 81.4% | Nemotron 3 Ultra |
| Long context reasoning (AA-LCR) | 69.0% | 79.3% | Nemotron 3 Ultra |
| DTBench | 90.1% | 90.1% | Tie |
| LMCA | 38.7% | 36.9% | Grok 4.20 |
| CL-bench | 22.2% | Not measured | Only Grok 4.20 measured |
| CL-bench Life | 11.9% | Not measured | Only Grok 4.20 measured |
| LMArena Text | Not measured | 1445 | Only Nemotron 3 Ultra measured |
| LMArena Text: instruction following | Not measured | 1418 | Only Nemotron 3 Ultra measured |
| LMArena Text: creative writing | Not measured | 1399 | Only Nemotron 3 Ultra measured |
| LMArena Text: English | Not measured | 1455 | Only Nemotron 3 Ultra measured |
| LMArena Text: non-English | Not measured | 1427 | Only Nemotron 3 Ultra measured |
| LMArena Text: Chinese | Not measured | 1497 | Only Nemotron 3 Ultra measured |
| LMArena Text: German | Not measured | 1472 | Only Nemotron 3 Ultra measured |
| LMArena Text: Korean | Not measured | 1384 | Only Nemotron 3 Ultra measured |
| LMArena Text: French | Not measured | 1472 | Only Nemotron 3 Ultra measured |
| LMArena Text: Spanish | Not measured | 1455 | Only Nemotron 3 Ultra measured |
| LMArena Text: Russian | Not measured | 1417 | Only Nemotron 3 Ultra measured |
| LMArena Text: Polish | Not measured | 1385 | Only Nemotron 3 Ultra measured |
| LMArena Text: multi-turn | Not measured | 1411 | Only Nemotron 3 Ultra measured |
| LMArena Text: longer queries | Not measured | 1437 | Only Nemotron 3 Ultra measured |
| LMArena Text: writing and language | Not measured | 1406 | Only Nemotron 3 Ultra measured |
| LMArena Text: entertainment and media | Not measured | 1397 | Only Nemotron 3 Ultra measured |
| Neither model has a result on the other 23 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Grok 4.20 | Nemotron 3 Ultra | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Nemotron 3 Ultra costs less
Repository review
50K input + 3K output tokens
Nemotron 3 Ultra costs less
Agent loop
200K input + 10K output tokens
Nemotron 3 Ultra costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | SpaceXAI Grok 4.20 | NVIDIA Nemotron 3 Ultra |
|---|---|---|
| Maker | xAI | NVIDIA |
| Weights | No public weights listed | Open weights |
| Released | March 31, 2026 | June 4, 2026 |
| Context window | 2M | 262K |
| Input $/Mtok | $1.25 | $0.60 |
| Output $/Mtok | $2.50 | $2.40 |
| Coding index | 33.3 | 30.0 |
| Independent measurements behind it | 2 | 3 |
| Cost per solved task | $0.72 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Grok 4.20 better than Nemotron 3 Ultra for coding?
Pick Grok 4.20 for coding: it scores 33.3 on the Vibe Coding Index against 30.0 for Nemotron 3 Ultra, though on 2 benchmarks against 3. Pick Nemotron 3 Ultra for output tokens 1.0x cheaper ($2.40 vs $2.50 per million) and work measured by LiveCodeBench (Vals AI) or Vibe Code Bench (Vals AI), where it leads.
Which is cheaper, Grok 4.20 or Nemotron 3 Ultra?
Grok 4.20 costs $1.25 per million input tokens and $2.50 per million output tokens; Nemotron 3 Ultra costs $0.60 and $2.40. Cost per solved task is known for Grok 4.20 only.
Which has the bigger context window, Grok 4.20 or Nemotron 3 Ultra?
Grok 4.20, with 2M tokens against 262K.
How many benchmarks is the Grok 4.20 vs Nemotron 3 Ultra comparison based on?
33 benchmarks have a result for both models as of October 22, 2026. Grok 4.20's coding index rests on 2 independent benchmarks and Nemotron 3 Ultra's on 3. Where only one model has a result, the page says so instead of naming a winner.