Nemotron 3 Super vs Qwen2.5 VL 72B Instruct
Nemotron 3 Super costs 2.2x less per output token than Qwen2.5 VL 72B Instruct ($0.45 vs $1.00 per million). Updated October 23, 2026; no benchmark has tested both models yet.
- Coding
- No data
- Overall
- No data
- Output $/M
- $0.45
- Coding
- No data
- Overall
- No data
- Output $/M
- $1.00
Nemotron 3 Super or Qwen2.5 VL 72B Instruct: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Nemotron 3 Super for output tokens 2.2x cheaper ($0.45 vs $1.00 per million) and a larger context window (262K against 128K tokens).
No benchmark has tested both yet. Nemotron 3 Super has results on AA Coding Index, SciCode and Terminal-Bench Hard (Artificial Analysis run); Qwen2.5 VL 72B Instruct on OSWorld, Kagi LLM Benchmark and GeoBench. 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
Nemotron 3 Super
Nemotron 3 Super takes 262K tokens against 128K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Nemotron 3 Super
Nemotron 3 Super $0.0003, Qwen2.5 VL 72B Instruct $0.0013 per turn.
List prices
At a glance
- Coding index
- Nemotron 3 Super not benchmarked, Qwen2.5 VL 72B Instruct 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
- Nemotron 3 Super $0.45 against Qwen2.5 VL 72B Instruct $1.00 per million output tokens, 2.2x apart.
- Context window
- Nemotron 3 Super takes 262K tokens, Qwen2.5 VL 72B Instruct 128K.
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 | Nemotron 3 Super | Qwen2.5 VL 72B Instruct | Leader |
|---|---|---|---|
| AA Coding Index | 37.7 | Not measured | Only Nemotron 3 Super measured |
| SciCode | 36.2% | Not measured | Only Nemotron 3 Super measured |
| Terminal-Bench Hard (Artificial Analysis run) | 28.8% | Not measured | Only Nemotron 3 Super measured |
| WeirdML | 38.0% | Not measured | Only Nemotron 3 Super measured |
| ALE-Bench | 213.9 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: coding | 1404 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: software and IT | 1409 | Not measured | 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 index1 board measured
| Benchmark | Nemotron 3 Super | Qwen2.5 VL 72B Instruct | Leader |
|---|---|---|---|
| AA Intelligence Index | 12.8 | Not measured | Only Nemotron 3 Super measured |
| Neither model has a result on the other 1 board here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index3 boards measured
| Benchmark | Nemotron 3 Super | Qwen2.5 VL 72B Instruct | Leader |
|---|---|---|---|
| AA Agentic Index | 1.7 | Not measured | Only Nemotron 3 Super measured |
| τ²-Bench Telecom | 67.8% | Not measured | Only Nemotron 3 Super measured |
| OSWorld | Not measured | 5.0% | Only Qwen2.5 VL 72B Instruct measured |
| Neither model has a result on the other 39 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index6 boards measured
| Benchmark | Nemotron 3 Super | Qwen2.5 VL 72B Instruct | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 80.0% | Not measured | Only Nemotron 3 Super measured |
| Humanity's Last Exam (Artificial Analysis run) | 20.8% | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: hard prompts | 1386 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: expert | 1398 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: hard prompts in English | 1397 | Not measured | Only Nemotron 3 Super measured |
| Kagi LLM Benchmark | Not measured | 36.0% | Only Qwen2.5 VL 72B Instruct measured |
| Neither model has a result on the other 17 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index3 boards measured
| Benchmark | Nemotron 3 Super | Qwen2.5 VL 72B Instruct | Leader |
|---|---|---|---|
| LMArena Text: math | 1380 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: mathematical work | 1405 | Not measured | Only Nemotron 3 Super measured |
| MathArena | 16.6% | Not measured | Only Nemotron 3 Super measured |
| Neither model has a result on the other 11 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index5 boards measured
| Benchmark | Nemotron 3 Super | Qwen2.5 VL 72B Instruct | Leader |
|---|---|---|---|
| GeoBench | Not measured | 62.0% | Only Qwen2.5 VL 72B Instruct measured |
| LMArena Text: science | 1388 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: medicine and healthcare | 1383 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: legal and government | 1392 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: business and finance | 1364 | Not measured | Only Nemotron 3 Super measured |
| Neither model has a result on the other 15 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index20 boards measured
| Benchmark | Nemotron 3 Super | Qwen2.5 VL 72B Instruct | Leader |
|---|---|---|---|
| IFBench | 71.5% | Not measured | Only Nemotron 3 Super measured |
| Long context reasoning (AA-LCR) | 65.7% | Not measured | Only Nemotron 3 Super measured |
| LMArena Text | 1378 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: instruction following | 1347 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: creative writing | 1317 | Not measured | Only Nemotron 3 Super measured |
| LMArena Vision | Not measured | 1107 | Only Qwen2.5 VL 72B Instruct measured |
| LMArena Text: English | 1400 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: non-English | 1356 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: Chinese | 1432 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: German | 1341 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: French | 1403 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: Spanish | 1418 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: Russian | 1336 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: Polish | 1346 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: multi-turn | 1369 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: longer queries | 1362 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: writing and language | 1331 | Not measured | Only Nemotron 3 Super measured |
| LMArena Text: entertainment and media | 1336 | Not measured | Only Nemotron 3 Super measured |
| LMArena Vision: English | Not measured | 1115 | Only Qwen2.5 VL 72B Instruct measured |
| LMArena Vision: Chinese | Not measured | 1107 | Only Qwen2.5 VL 72B Instruct measured |
| Neither model has a result on the other 28 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 | NVIDIA Nemotron 3 Super | Qwen2.5 VL 72B Instruct |
|---|---|---|
| Maker | NVIDIA | Qwen |
| Weights | Open weights | Open weights |
| Released | March 11, 2026 | February 1, 2025 |
| Context window | 262K | 128K |
| Input $/Mtok | $0.085 | $0.80 |
| Output $/Mtok | $0.45 | $1.00 |
| 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 Nemotron 3 Super better than Qwen2.5 VL 72B Instruct for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Nemotron 3 Super for output tokens 2.2x cheaper ($0.45 vs $1.00 per million) and a larger context window (262K against 128K tokens).
Which is cheaper, Nemotron 3 Super or Qwen2.5 VL 72B Instruct?
Nemotron 3 Super costs $0.085 per million input tokens and $0.45 per million output tokens; Qwen2.5 VL 72B Instruct costs $0.80 and $1.00. Cost per solved task is known for neither of them.
Which has the bigger context window, Nemotron 3 Super or Qwen2.5 VL 72B Instruct?
Nemotron 3 Super, with 262K tokens against 128K.
Has any benchmark tested both Nemotron 3 Super and Qwen2.5 VL 72B Instruct?
No benchmark has tested both yet. Nemotron 3 Super has results on AA Coding Index, SciCode and Terminal-Bench Hard (Artificial Analysis run); Qwen2.5 VL 72B Instruct on OSWorld, Kagi LLM Benchmark and GeoBench. Compare them on price and context instead. Data as of October 23, 2026.