Nemotron 3 Ultra
by NVIDIA · nvidia/nemotron-3-ultra-550b-a55b · #233 cheapest of 457 paid models
Prices updated Nov 18, 2026, 3:51 PM UTC · refreshed nightly
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...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.50
per 1M tokens
Output
$2.20
per 1M tokens
Blended (3:1)
$0.925
3 input : 1 output
Cache read
$0.10
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for Nemotron 3 Ultra with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.925 | $0.0532 | — |
| 50% | $0.775 | $0.0332 | −38% |
| 90% | $0.655 | $0.0172 | −68% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.00091 | $0.91 |
| Document summary | 8,000 | 1,000 | $0.0062 | $6.20 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.0194 | $19.40 |
| Long-context analysis | 150,000 | 5,000 | $0.086 | $86.00 |
Performance
Intelligence Index
11.8
AA composite quality
Coding Index
49.3
Math Index
—
Agentic Index
20.1
Output speed
—
median
Time to first token
—
median
Time to first answer
—
after reasoning tokens
Design Arena Elo
| Category | Elo |
|---|---|
| Website | 1,128 |
| UI components | 1,125 |
| Data viz | 1,137 |
| SVG | 1,060 |
| 3D | 1,109 |
| Game dev | 1,126 |
| Code categories | 1,130 |
| ASCII art | 1,079 |
Specs
Context window
262K
tokens
Max output
16K
tokens
Input modalities
text
Tokenizer
Other