Nemotron 3 Ultra 550B A55B
nvidia/nemotron-3-ultra-550b-a55bLargest Nemotron 3 model for maximum open-weight reasoning and agent accuracy
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...
Context
1M
1,000,000 tokens
Max output
66K
65,536 tokens
Input
Free
per 1M tokens
Output
Free
per 1M tokens
Providers
23
29 listings
Intelligence
22.9
Intelligence Index
Pricing
USD per 1M tokens · listed by Nvidia
| Token type | Standard |
|---|---|
| Input | Free |
| Output | Free |
Lowest input price across providers: Free.
Capabilities
ReasoningYes
Effort: none, medium, high · Budget tokens ≥ -1, ≤ 32,768
Tool callYes
Structured outputYes
TemperatureYes
AttachmentsNo
Interleaved reasoningYes
Field: reasoning_content
Open weightsYes
ModeratedNo
Input
Text
Output
Text
Benchmarks
Published scores and Design Arena Elo
- Intelligence22.9#85 of 208
- Coding49.3#57 of 112
- Agentic20.1#42 of 86
- Arena1,125#83 of 128
Rankings
Position on each OneAtlas leaderboard
Providers · 23
29 listings
| Provider | Model ID | Context | Input limit | Output | Input $ | Output $ | Reasoning $ | Cache read | Cache write | Audio in | Audio out | Reasoning | Tools | Structured | Temp. | Updated |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| nemotron-3-ultra-550b-a55b-free | 1M | — | 128K | Free | Free | — | — | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B | 203K | — | 203K | $0.60 | $2.40 | — | $0.12 | — | — | — | Yes | Yes | Yes | Yes | Jun 4, 2026 | |
| nemotron-3-ultra-550b-a55b:free | 1M | — | 128K | Free | Free | — | — | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia/nemotron-3-ultra-550b-a55b | 1M | — | 128K | $0.60 | $2.40 | — | $0.12 | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B | 262K | — | 262K | $0.50 | $2.15 | — | $0.10 | — | — | — | Yes | Yes | Yes | Yes | Jun 4, 2026 | |
| nemotron-3-ultra-550b | 1.05M | — | 128K | $0.50 | $2.20 | — | $0.10 | — | — | — | Yes | Yes | No | Yes | Jun 4, 2026 | |
| nemotron-3-ultra-550b | 131K | — | 131K | $0.90 | $1.70 | — | $0.18 | — | — | — | No | Yes | Unknown | Yes | Jun 4, 2026 | |
| arkanecloud/nvidia/nemotron-3-ultra-550b-a55b | 512K | — | 128K | $0.60 | $3.50 | — | $0.20 | — | — | — | Yes | Yes | No | Yes | Jun 4, 2026 | |
| deepinfra/nemotron-3-ultra-550b-a55b | 262K | — | 128K | $0.50 | $2.20 | — | $0.10 | — | — | — | Yes | Yes | Yes | Yes | Jun 4, 2026 | |
| nebius/nvidia/Nemotron-3-Ultra-550b-a55b | 1.05M | — | 128K | $1.00 | $3.00 | — | $1.00 | — | — | — | Yes | Yes | No | Yes | Jun 4, 2026 | |
| accounts/fireworks/models/nemotron-3-ultra-nvfp4 | 262K | — | 128K | $0.60 | $2.40 | — | $0.12 | — | — | — | Yes | Yes | Yes | Yes | Jun 4, 2026 | |
| nemotron-3-ultra-550b-a55b | 1M | — | 128K | $0.69 | $4.29 | — | $0.07 | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia/nemotron-3-ultra-550b-a55b | 203K | — | 183K | $0.40 | $1.76 | — | $0.08 | — | — | — | Yes | Yes | Yes | Yes | Jun 4, 2026 | |
| nvidia/nemotron-3-ultra-550b-a55b:free | 1M | — | 66K | Free | Free | — | — | — | — | — | Yes | Yes | No | Yes | Jun 4, 2026 | |
| nvidia/nemotron-3-ultra-550b-a55b | 1M | 1M | 66K | $0.50 | $2.50 | — | $0.25 | — | — | — | Yes | Yes | No | Yes | Jun 4, 2026 | |
| nvidia/nemotron-3-ultra-550b-a55b:thinking | 1M | 1M | 66K | $0.50 | $2.50 | — | $0.25 | — | — | — | Yes | Yes | No | Yes | Jun 4, 2026 | |
| nvidia/Nemotron-3-Ultra-550b-a55b | 1.05M | — | 1.05M | $1.00 | $3.00 | — | $0.12 | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia/nemotron-3-ultra-550b-a55b | 1M | — | 66K | Free | Free | — | — | — | — | — | Yes | Yes | Yes | Yes | Jun 4, 2026 | |
| nemotron-3-ultra | 262K | — | 128K | $0.10 | $3.00 | — | $0.10 | — | — | — | Yes | Yes | No | Yes | Jun 4, 2026 | |
| nemotron-3-ultra-free | 1M | — | 128K | Free | Free | — | Free | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| Requesty | nemotron-3-ultra-550b-a55b | 1.05M | — | 66K | Free | Free | — | — | — | — | — | Yes | Yes | No | Yes | Jun 4, 2026 |
| Requesty | nvidia-nemotron-3-ultra | 262K | — | 131K | $0.50 | $2.50 | — | — | — | — | — | Yes | No | No | Yes | Jun 4, 2026 |
| nemotron-3-ultra | 131K | — | 32K | $0.10 | $0.10 | — | — | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 | 66K | — | 66K | $1.80 | $6.22 | — | $0.36 | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16:peft:262144 | 262K | — | 262K | $1.80 | $6.22 | — | $0.36 | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia/nemotron-3-ultra-550b-a55b | 512K | — | 512K | $0.60 | $3.60 | — | $0.20 | — | — | — | Yes | Yes | Yes | Yes | Jun 4, 2026 | |
| nemotron-3-ultra-550b-a55b:free | 1M | — | 128K | Free | Free | — | — | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 | |
| nvidia-nemotron-3-ultra-550b-a55b | 256K | — | 33K | $0.63 | $3.13 | — | $0.19 | — | — | — | Yes | Yes | Yes | Yes | Jun 4, 2026 | |
| nvidia/nemotron-3-ultra-550b-a55b | 1M | — | 65K | $0.60 | $2.40 | — | $0.12 | — | — | — | Yes | Yes | Unknown | Yes | Jun 4, 2026 |