DeepSeek-R1 vs NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16
Specs, VRAM requirements and download trends — updated 24 October 2026.
deepseek-ai · A · Open weights
Params
684.5 B
Context
163,840
30d
1.4 M
mit · commercial OK
nvidia · B · Open weights
Params
560.5 B
Context
262,144
30d
1.1 M
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | DeepSeek-R1 | NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 684.5 B | 560.5 B |
| Architecture | DeepseekV3ForCausalLM | NemotronHForCausalLM |
| Context length | 163,840 | 262,144 |
| Licence | mit | other |
| Commercial use | Allowed | Unknown |
| Languages | — | 12 |
| Downloads 30d | 1,391,551 | 1,067,855 |
| Downloads all time | 50.3 M | 2.1 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2025-01-20 | 2026-06-03 |
| API input price / 1M tokens | $0.70 | $0.60 |
| API output price / 1M tokens | $2.50 | $2.40 |
| Est. cost per 1M input + 1M output tokens | $3.20 | $3.00 |
| Cached input / 1M tokens | — | $0.12 |
| AA Intelligence Index | 11.4 | 22.9 |
| AA Coding Index | 24.6 | 49.3 |
| AA Agentic Index | 1.1 | 20.1 |
| Max output tokens | 16,000 | 182,520 |
| API providers | — | 3 |
| Tool calling | Yes | Yes |
| Reasoning mode | Yes | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
DeepSeek-R1
$3.20
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16
$3.00
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 is 1.1× cheaper at list price.
Download trend
Daily snapshots, last 89 days (28 Jul – 24 Oct)
9.9 M213 K
DeepSeek-R1
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
DeepSeek-R1 · fp16 @ 160K ctx
2,811.5 GB / 24 GB
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 · fp16 @ 256K ctx
3,924.2 GB / 24 GB