NVIDIA: Nemotron 3 Ultra
nvidia/nemotron-3-ultra-550b-a55bNVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).
Modalities
In / out per 1M$0.60 / $2.40
Context256K
ReleasedJun 4, 2026
Providers
4 providers serve this model upstream.
| Provider | Input /M | Output /M | Cache read /M | Latency P50 | Throughput P50 | Uptime |
|---|---|---|---|---|---|---|
 DeepInfraus · fp4 | $0.50 | $2.20 | $0.10 | 1.75s | 46 tps | 100% |
 Basetenus · fp4 | $0.60 | $2.40 | $0.12 | 0.46s | 85 tps | 100% |
| Basetenfp4 | $0.60 | $2.40 | $0.12 | 0.41s | 84 tps | 100% |
| $0.63 | $3.13 | $0.19 | 1.19s | 39 tps | 99.24% |
Market-wide measurements, refreshed every 15 minutes. They describe who serves Nemotron 3 Ultra in general, not how ProjectCOZY routes your requests.
Pricing
List rates per 1M tokens — what Nemotron 3 Ultra costs pay-as-you-go elsewhere, not what you pay here.
Input
$0.60
per 1M tokens
Output
$2.40
per 1M tokens
Cache read
$0.12
per 1M cached tokens
Included in every pass
Nemotron 3 Ultra is included in every ProjectCOZY pass at no extra cost — one flat rate, no per-token metering. The rates above are there for comparison.
Performance
Time to first token and output speed across every provider, over the last 3 days.
0.0s2.0s4.0s6.0s8.0s
Oct 11, 12 AMOct 11, 8 AMOct 11, 4 PMOct 12, 12 AMOct 12, 8 AMOct 12, 4 PMOct 13, 12 AMOct 13, 8 AMOct 13, 4 PMOct 14, 12 AMOct 14, 8 AM
Venice DeepInfra (us) Baseten Baseten (us)
0 tok/s200 tok/s400 tok/s600 tok/s800 tok/s
Oct 11, 12 AMOct 11, 8 AMOct 11, 4 PMOct 12, 12 AMOct 12, 8 AMOct 12, 4 PMOct 13, 12 AMOct 13, 8 AMOct 13, 4 PMOct 14, 12 AMOct 14, 8 AM
Venice DeepInfra (us) Baseten Baseten (us)
Uptime
100% of requests were served successfully over the last 3 days, across all providers.
Last 3 daysAvailability 99.9%
Oct 11, 2:00 PMNow
Availability over the last 24 hoursAvailability 99.89%
100% 98% 97% 95%
Oct 13, 2:40 PMNow
Apps
Tools that most often run Nemotron 3 Ultra, by tokens this week.
Activity
134.2B tokens processed market-wide in the last month.
02.0B4.0B6.0B8.0B
Sep 14Sep 19Sep 24Sep 29Oct 4Oct 9Oct 14
Tokens
FAQ
Which model id do I use for Nemotron 3 Ultra?
nvidia/nemotron-3-ultra-550b-a55b. Send it as the model field from any OpenAI-compatible client, against http://192.168.1.55/v1.
How much context does Nemotron 3 Ultra hold?
262,144 tokens per request, of which up to 182,520 can be the response.
What can it accept and return?
Input: text. Output: text.
What does Nemotron 3 Ultra cost on a pass?
Nothing beyond the pass itself. Pay-as-you-go elsewhere it lists at $0.60 per 1M input tokens and $2.40 per 1M output tokens.
Who built Nemotron 3 Ultra?
NVIDIA, in the Nemotron family.
Explore
Other models your pass can call.
