NVIDIA · Updated 4 November 2026
Nemotron 3 Ultra 550B A55B
Nemotron 3 Ultra 550B A55B from NVIDIA is #43 of 54 models on the Artificial Analysis Intelligence Index, with a score of 12. It costs $1.13 per million tokens ($0.60 in, $2.70 out) and writes 161 tok/s. Its weights are open: anyone can download and run it.
- Intelligence
- 12 · 43rd of 54
- Coding
- Not scored yet
- Price per 1M tokens
- $1.13 blended · $0.60 input · $2.70 output
- Output speed
- 161 tok/s
- Time to first token
- 0.9 s
- Context window
- 262K tokens
- Open weights
- Yes
- Released
- 4 June 2026
How it compares
| # | Model | Intelligence | Coding | Price / 1M tokens | Speed | Context |
|---|---|---|---|---|---|---|
| 1 | Nemotron 3 Ultra 550B A55B NVIDIA · open weights | 12 | — | $1.13 | 161 tok/s | 262K |
| 2 | Gemini 4 Carbon Google | 60 | — | $8.00 | 92 tok/s | — |
| 3 | Inkling Small Thinking Machines · open weights | 13 | — | $0.53 | 201 tok/s | 524K |
| 4 | Inkling Thinking Machines · open weights | 13 | — | $1.76 | 196 tok/s | 524K |
| 5 | Gemini 3.5 Flash-Lite Google | 12 | — | $0.85 | 374 tok/s | 1M |
| 6 | MiMo-V2.6-Flash Xiaomi · open weights | 27 | — | $0.17 | 30 tok/s | 1.1M |
Cheaper alternative within 3 points: MiMo-V2.6-Flash at $0.17 per million tokens.
Data: Artificial Analysis (intelligence and coding indexes, prices, speed) and OpenRouter (open weights, context windows). Updated 4 November 2026; refreshed every 6 hours.
More rankings
All rankings · Best overall · Best open model · Best value · Cheapest good model · Best for coding · Fastest · Longest context · Price table