Nemotron 3 Ultra vs Nemotron 3.5 Content Safety
Nemotron 3 Ultra (NVIDIA) and Nemotron 3.5 Content Safety (NVIDIA) are listed at different prices. Nemotron 3.5 Content Safety costs less: $0.20 per 1M input tokens and $0.20 per 1M output tokens, against $0.50 and $2.20 for Nemotron 3 Ultra. Nemotron 3 Ultra has the larger context window, 262,144 tokens against 131,072 for Nemotron 3.5 Content Safety. Only Nemotron 3 Ultra supports tool calling. Nemotron 3 Ultra accepts text and produces text; Nemotron 3.5 Content Safety accepts text and images and produces text.
NVIDIA: Nemotron 3 Ultra accepts text and supports tools, structured outputs, response formats, and reasoning; NVIDIA: Nemotron 3.5 Content Safety accepts text and images. Both produce text and support reasoning, while their context limits are 262,144 and 131,072 tokens, respectively.
Compared from catalog data checked
Side by side
0.00 the better figure in a row| Specification | Nemotron 3 Ultra | Nemotron 3.5 Content Safety |
|---|---|---|
| Maker | NVIDIA | NVIDIA |
| Added to catalog | 4 Jun 2026 | 4 Jun 2026 |
| Input, per 1M tokens | $0.50 | $0.20 |
| Output, per 1M tokens | $2.20 | $0.20 |
| Cached input read, per 1M | $0.10 | n/a |
| Context windowTokens | 262,144 | 131,072 |
| Max outputTokens | 16,384 | 117,964 |
| Accepts | text | image, text |
| Produces | text | text |
| Tool calling | Yes | No |
| Structured outputs | Yes | No |
| Reasoning effort control | Yes | No |
List prices through OpenRouter, in US dollars. Each model's page has its variants, price history and sample outputs.
What the same task costs
List price for one request, from the token counts shown. The same four tasks are priced on every model page.
| Task | Nemotron 3 Ultra | Nemotron 3.5 Content Safety |
|---|---|---|
| Answer a short question500 in / 300 out | $0.0009 | $0.0002 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.014 | $0.0052 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.019 | $0.0064 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.078 | $0.030 |
Which to choose
Choose Nemotron 3 Ultra if
- Cached input costs less: $0.10 against n/a per 1M
- You need the larger context window: 262,144 tokens against 131,072
- You need tool calling
- You need answers in a fixed JSON schema
- You want to set how much it reasons before it answers
Choose Nemotron 3.5 Content Safety if
- Input tokens cost less: $0.20 against $0.50 per 1M
- Output tokens cost less: $0.20 against $2.20 per 1M
- You need longer answers in one response: 117,964 tokens against 16,384
- It has to read images, which the other model does not
Neither maker's quality claims are repeated here. The reliable test is your own task: ask both the same question and read the two answers.