NVIDIALlama Nemotron Super 49B v1.5
Not yet ranked: too few independent results so far. Released 25 Jul 2025.
- Blended price
- $0.400/M
- $0.400 in · $0.400 out
- Output speed
- 58 tok/s
- First answer
- 44 s
- Context
- 128k
- Agents & tools38
- Knowledge42
- Instruction following40
- Composite40
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | thinking | Source |
|---|---|---|
| GPQA Diamond (AA)not in index | 74.8%#245 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 7.3%#298 | Artificial Analysis |
Agents & tools
| Benchmark | thinking | Source |
|---|---|---|
| Terminal-Bench Hard | 5.3%#263 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 28.1%#261 | Artificial Analysis |
Knowledge
| Benchmark | thinking | Source |
|---|---|---|
| AA-Omniscience | -47.6#296 | Artificial Analysis |
Instruction following
| Benchmark | thinking | Source |
|---|---|---|
| IFBench | 37.0%#280 | Artificial Analysis |
Composite
| Benchmark | thinking | Source |
|---|---|---|
| AA Intelligence Index | 9.0#340 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0003 | – | 49.0 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0050 | – | 54.1 s |
| Code edit | 6,000 / 1,500 | $0.0030 | – | 1.2 min |
| Agentic coding session | 60,000 / 4,000 | $0.026 | – | 1.9 min |
| Structured extraction | 2,000 / 200 | $0.0009 | – | 47.2 s |
See also
Data as of 9 Sept 2026. Compare with another model.