Nemotron 3 Ultra
The 9th most-used model on the router over the last 7 days, with 3.1% of tokens. Weekly share peaked at 5.1% in the week of Aug 17, 2026, with 69T tokens processed since Jun 1, 2026.
- Launch
- Jun 4, 2026
- Context
- 256k tokens
- Modality
- text → text
- Reasoning
- yes
- License
- Open weights
- Input
- $0.60 / 1M
- Output
- $2.40 / 1M
Weekly token share
Share of the router's weekly volume, since the first week with volume
See the numbers
| Week | Share |
|---|---|
| Oct 19, 2026 | 3.112% |
| Oct 12, 2026 | 3.029% |
| Oct 5, 2026 | 4.000% |
| Sep 28, 2026 | 3.406% |
| Sep 21, 2026 | 3.891% |
| Sep 14, 2026 | 3.487% |
| Sep 7, 2026 | 2.809% |
| Aug 31, 2026 | 3.149% |
| Aug 24, 2026 | 4.713% |
| Aug 17, 2026 | 5.084% |
| Aug 10, 2026 | 2.847% |
| Aug 3, 2026 | 3.486% |
| Jul 27, 2026 | 4.274% |
| Jul 20, 2026 | 4.031% |
| Jul 13, 2026 | 5.011% |
| Jul 6, 2026 | 3.923% |
| Jun 29, 2026 | 2.206% |
| Jun 22, 2026 | 1.530% |
| Jun 15, 2026 | 1.348% |
| Jun 8, 2026 | 1.780% |
| Jun 1, 2026 | 0.823% |
Debuted in the week of Jun 1, 2026 and peaked at 5.1% 11 weeks later. The latest full week came in at 3.1%, 61% of peak.
What this model is used for new
Tasks where the model ranks among the leaders, and its share of each
Ranks among the leaders in 6 of the 29 classified tasks. Its largest share is in frontend and UI, 9.7% of the task's tokens; in the heaviest task where it appears, code generation (9.0% of classified volume), it holds 7.3%.
Snapshot of the rolling 7-day window through Oct 25, 2026. Share: the model's tokens in the task over the task's total tokens. Weight: the task's share of classified volume.
Where to run it
Endpoints in the source's catalog, from cheapest to most expensive on input
| Provider | Input | Output | Context | Quantization | Zero retention |
|---|---|---|---|---|---|
| DeepInfra | $0.40 | $1.76 | 256k | fp4 | yes |
| DeepInfra | $0.50 | $2.20 | 256k | fp4 | yes |
| BaseTen | $0.60 | $2.40 | 198k | fp4 | yes |
| BaseTen | $0.60 | $2.40 | 198k | fp4 | yes |
| Venice | $0.625 | $3.13 | 250k | fp8 | yes |
5 endpoints across 3 providers, with input from $0.40 to $0.625 per 1M tokens. 5 offer zero data retention.
Prices in US dollars per million tokens, from the latest archived endpoint catalog.
Benchmarks
Artificial Analysis indexes and benchmarks run by the source, with cost per task
GPQA Diamond 81.2%
$0.14 per task, 172 tasks. 99th of 169 evaluated. Median of evaluated models: 83.4%, $0.025 per task.
τ-bench verified, airline 72.0%
$0.15 per task, 50 tasks. 66th of 150 evaluated. Median of evaluated models: 71.2%, $0.064 per task.
Artificial Analysis indexes, exposed by the catalog; benchmarks run by the source and archived on Oct 26, 2026. The gray mark is the median of evaluated models.
Against the week's largest models
The five largest proprietary and five largest open-weights models over the last 7 days
| Model | Lab | 7-day share | Blended price | Context | Intelligence | Weeks in top 10 |
|---|---|---|---|---|---|---|
| Proprietary | ||||||
| GPT-6 LunaOpenAI | OpenAI | 4.4% | $0.20 | 1M | 38.1 | 5 |
| Claude Opus 5.5Anthropic | Anthropic | 4.4% | $8.00 | 1M | 57.6 | 3 |
| jev-1.13Typesafe | Typesafe | 1.6% | — | — | — | 0 |
| GPT-6.1 SolOpenAI | OpenAI | 1.5% | $4.00 | 1M | 51.8 | 0 |
| Muse Spark 1.3 ContributorMeta | Meta | 1.5% | $0.125 | 1M | — | 0 |
| Open weights | ||||||
| DeepSeek V4.1 Flash 1017DeepSeek | DeepSeek | 17.9% | $0.263 | 1M | 39.5 | 1 |
| DeepSeek V4.1 FlashDeepSeek | DeepSeek | 9.6% | $0.265 | 1M | 24.7 | 7 |
| MiniMax M3.1 FlashMiniMax | MiniMax | 7.7% | $0.263 | 1M | 42.7 | 1 |
| GLM 5.3 FlashZ.ai (GLM) | Z.ai (GLM) | 7.0% | $0.237 | 1M | 41.8 | 9 |
| Hy4 previewTencent | Tencent | 4.8% | $1.25 | 1M | — | 8 |
| Nemotron 3 Ultrathis pageNVIDIA | NVIDIA | 3.1% | $1.05 | 256k | 22.9 | 16 |
Share over the last 7 days (Oct 19, 2026 to Oct 25, 2026). Blended price in US dollars per million tokens, weighted 75% input and 25% output. A dash means the number is not published, not zero.