GPT-5.2 Pro vs Nemotron 3 Super
Nemotron 3 Super costs 373.3x less per output token than GPT-5.2 Pro ($0.45 vs $168.00 per million). Updated October 22, 2026; no benchmark has tested both models yet.
- Coding
- No data
- Overall
- No data
- Output $/M
- $168.00
- Coding
- No data
- Overall
- No data
- Output $/M
- $0.45
GPT-5.2 Pro or Nemotron 3 Super: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick GPT-5.2 Pro for a larger context window (400K against 262K tokens).
No benchmark has tested both yet. GPT-5.2 Pro has results on Epoch Capabilities Index, ARC-AGI and ARC-AGI-2; Nemotron 3 Super on AA Coding Index, SciCode and Terminal-Bench Hard (Artificial Analysis run). Compare them on price and context instead.
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
No call
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
No call
Only Nemotron 3 Super has a AA Intelligence Index result, 12.8.
Directional only
Agentic work
AA Agentic Index
No call
Only Nemotron 3 Super has a AA Agentic Index result, 1.7.
Directional only
Long documents
Listed context window
GPT-5.2 Pro
GPT-5.2 Pro takes 400K tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Nemotron 3 Super
GPT-5.2 Pro $0.105, Nemotron 3 Super $0.0003 per turn.
List prices
At a glance
- Coding index
- GPT-5.2 Pro not benchmarked, Nemotron 3 Super not benchmarked.
- Widest category
- None yet: no category has been tested on both models.
- Widest single gap
- None yet: no benchmark has tested both models.
- Output price
- GPT-5.2 Pro $168.00 against Nemotron 3 Super $0.45 per million output tokens, 373.3x apart.
- Context window
- GPT-5.2 Pro takes 400K tokens, Nemotron 3 Super 262K.
Benchmark by benchmark
Every board either model has a result on, a category to a fold; bars are each result on a 0 to 100 scale. Only one model is measured on each, so no row names a leader.
CodingDirectional onlySix of these boards feed the Vibe Coding Index7 boards measured
| Benchmark | GPT-5.2 Pro | Nemotron 3 Super | Leader |
|---|---|---|---|
| AA Coding Index | Not measured | 37.7 | Only Nemotron 3 Super measured |
| SciCode | Not measured | 36.2% | Only Nemotron 3 Super measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 28.8% | Only Nemotron 3 Super measured |
| WeirdML | Not measured | 38.0% | Only Nemotron 3 Super measured |
| ALE-Bench | Not measured | 213.9 | Only Nemotron 3 Super measured |
| LMArena Text: coding | Not measured | 1403 | Only Nemotron 3 Super measured |
| LMArena Text: software and IT | Not measured | 1409 | Only Nemotron 3 Super measured |
| Neither model has a result on the other 36 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards measured
| Benchmark | GPT-5.2 Pro | Nemotron 3 Super | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 12.8 | Only Nemotron 3 Super measured |
| Epoch Capabilities Index | 156.3 | Not measured | Only GPT-5.2 Pro measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards measured
| Benchmark | GPT-5.2 Pro | Nemotron 3 Super | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 1.7 | Only Nemotron 3 Super measured |
| τ²-Bench Telecom | Not measured | 67.8% | Only Nemotron 3 Super measured |
| Neither model has a result on the other 40 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index9 boards measured
| Benchmark | GPT-5.2 Pro | Nemotron 3 Super | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 80.0% | Only Nemotron 3 Super measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 20.8% | Only Nemotron 3 Super measured |
| ARC-AGI | 90.5% | Not measured | Only GPT-5.2 Pro measured |
| ARC-AGI-2 | 54.2% | Not measured | Only GPT-5.2 Pro measured |
| SimpleBench | 57.4% | Not measured | Only GPT-5.2 Pro measured |
| Chess puzzles | 60.0% | Not measured | Only GPT-5.2 Pro measured |
| LMArena Text: hard prompts | Not measured | 1386 | Only Nemotron 3 Super measured |
| LMArena Text: expert | Not measured | 1398 | Only Nemotron 3 Super measured |
| LMArena Text: hard prompts in English | Not measured | 1397 | Only Nemotron 3 Super measured |
| Neither model has a result on the other 14 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index5 boards measured
| Benchmark | GPT-5.2 Pro | Nemotron 3 Super | Leader |
|---|---|---|---|
| FrontierMath, tiers 1 to 3 | 74.0% | Not measured | Only GPT-5.2 Pro measured |
| FrontierMath, tier 4 | 46.0% | Not measured | Only GPT-5.2 Pro measured |
| LMArena Text: math | Not measured | 1380 | Only Nemotron 3 Super measured |
| LMArena Text: mathematical work | Not measured | 1405 | Only Nemotron 3 Super measured |
| MathArena | Not measured | 16.6% | Only Nemotron 3 Super measured |
| Neither model has a result on the other 9 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index6 boards measured
| Benchmark | GPT-5.2 Pro | Nemotron 3 Super | Leader |
|---|---|---|---|
| LMArena Text: science | Not measured | 1387 | Only Nemotron 3 Super measured |
| LMArena Text: medicine and healthcare | Not measured | 1383 | Only Nemotron 3 Super measured |
| LMArena Text: legal and government | Not measured | 1392 | Only Nemotron 3 Super measured |
| LMArena Text: business and finance | Not measured | 1364 | Only Nemotron 3 Super measured |
| PRBench Finance | 46.3% | Not measured | Only GPT-5.2 Pro measured |
| PRBench Legal | 45.4% | Not measured | Only GPT-5.2 Pro measured |
| Neither model has a result on the other 14 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index17 boards measured
| Benchmark | GPT-5.2 Pro | Nemotron 3 Super | Leader |
|---|---|---|---|
| IFBench | Not measured | 71.5% | Only Nemotron 3 Super measured |
| Long context reasoning (AA-LCR) | Not measured | 65.7% | Only Nemotron 3 Super measured |
| LMArena Text | Not measured | 1378 | Only Nemotron 3 Super measured |
| LMArena Text: instruction following | Not measured | 1347 | Only Nemotron 3 Super measured |
| LMArena Text: creative writing | Not measured | 1317 | Only Nemotron 3 Super measured |
| LMArena Text: English | Not measured | 1400 | Only Nemotron 3 Super measured |
| LMArena Text: non-English | Not measured | 1356 | Only Nemotron 3 Super measured |
| LMArena Text: Chinese | Not measured | 1431 | Only Nemotron 3 Super measured |
| LMArena Text: German | Not measured | 1341 | Only Nemotron 3 Super measured |
| LMArena Text: French | Not measured | 1403 | Only Nemotron 3 Super measured |
| LMArena Text: Spanish | Not measured | 1418 | Only Nemotron 3 Super measured |
| LMArena Text: Russian | Not measured | 1336 | Only Nemotron 3 Super measured |
| LMArena Text: Polish | Not measured | 1346 | Only Nemotron 3 Super measured |
| LMArena Text: multi-turn | Not measured | 1369 | Only Nemotron 3 Super measured |
| LMArena Text: longer queries | Not measured | 1362 | Only Nemotron 3 Super measured |
| LMArena Text: writing and language | Not measured | 1332 | Only Nemotron 3 Super measured |
| LMArena Text: entertainment and media | Not measured | 1336 | Only Nemotron 3 Super measured |
| Neither model has a result on the other 31 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 Super costs less
Repository review
50K input + 3K output tokens
Nemotron 3 Super costs less
Agent loop
200K input + 10K output tokens
Nemotron 3 Super 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 | OpenAI GPT-5.2 Pro | NVIDIA Nemotron 3 Super |
|---|---|---|
| Maker | OpenAI | NVIDIA |
| Weights | No public weights listed | Open weights |
| Released | December 10, 2025 | March 11, 2026 |
| Context window | 400K | 262K |
| Input $/Mtok | $21.00 | $0.085 |
| Output $/Mtok | $168.00 | $0.45 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.2 Pro better than Nemotron 3 Super for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick GPT-5.2 Pro for a larger context window (400K against 262K tokens).
Which is cheaper, GPT-5.2 Pro or Nemotron 3 Super?
GPT-5.2 Pro costs $21.00 per million input tokens and $168.00 per million output tokens; Nemotron 3 Super costs $0.085 and $0.45. Cost per solved task is known for neither of them.
Which has the bigger context window, GPT-5.2 Pro or Nemotron 3 Super?
GPT-5.2 Pro, with 400K tokens against 262K.
Has any benchmark tested both GPT-5.2 Pro and Nemotron 3 Super?
No benchmark has tested both yet. GPT-5.2 Pro has results on Epoch Capabilities Index, ARC-AGI and ARC-AGI-2; Nemotron 3 Super on AA Coding Index, SciCode and Terminal-Bench Hard (Artificial Analysis run). Compare them on price and context instead. Data as of October 22, 2026.