Nemotron API Providers
Who serves Nemotron, at what price and speed.
Nemotron is NVIDIA's open-weight model family, from Nano to the 550B-parameter Nemotron 3 Ultra. NVIDIA's own free variants carry most of the volume.
NVIDIA served 89% of the 8.16T Nemotron tokens that went through a large public LLM router over Nov 2 to Nov 8, 2026. NVIDIA's own API served 89%; 12 providers list the family.
Who serves Nemotron tokens
All Nemotron models combined, Nov 2 to Nov 8, 2026.
Nemotron 3.5 Super VL: prices and speed by provider
Output prices run from $0.30 to $0.90 per million tokens across 10 providers, median $0.50.
| Provider | Share | Input $/M | Output $/M | Cache read $/M | Tokens/s | Latency | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| CoreWeave | 16.5% | $0.15 | $0.50 | $0.02 | 297 | 0.48s | 99.3% | 1.05M |
| Nebius Token Factory | 2.8% | $0.30 | $0.90 | $0.06 | 77 | 1.95s | 99.5% | 1.05M |
| Together | 1.8% | $0.15 | $0.50 | $0.02 | 150 | 0.70s | 98.6% | 262K |
| Prime Intellect | 1.1% | $0.15 | $0.50 | $0.02 | 162 | 3.55s | 100.0% | 1.05M |
| Crusoe | 0.7% | $0.10 | $0.40 | $0.08 | 47 | 2.28s | 96.3% | 1.05M |
| Venice | 0.2% | $0.20 | $0.50 | $0.04 | 343 | 0.46s | 98.8% | 262K |
| io.net | 0.2% | $0.09 | $0.30 | $0.012 | 33 | 27.16s | 99.9% | 262K |
| Baseten | <0.1% | $0.20 | $0.50 | $0.04 | 190.5 | 1.65s | 99.9% | 1.05M |
| Darkbloom | <0.1% | $0.06 | $0.35 | $0.01 | 24 | 3.82s | 88.6% | 262K |
| DeepInfra | - | $0.15 | $0.50 | $0.02 | 168 | 1.83s | 100.0% | 262K |
List prices as of November 8, 2026, cheapest endpoint per provider. Tokens/s and latency are medians over 30 minutes.
Nemotron 3 Ultra: prices and speed by provider
Output prices run from $1.76 to $3.13 per million tokens across 3 providers, median $2.40.
| Provider | Share | Input $/M | Output $/M | Cache read $/M | Tokens/s | Latency | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| Baseten | 0.6% | $0.60 | $2.40 | $0.12 | 114 | 0.25s | 99.9% | 203K |
| Venice | 0.1% | $0.625 | $3.13 | $0.188 | 70 | 0.81s | 99.9% | 256K |
| DeepInfra | - | $0.40 | $1.76 | $0.08 | 51 | 1.36s | 100.0% | 262K |
List prices as of November 8, 2026, cheapest endpoint per provider. Tokens/s and latency are medians over 30 minutes.
All Nemotron models
| Model | Tokens, 7d | Providers | Output $/M | Context | Released |
|---|---|---|---|---|---|
| Nemotron 3.5 Super VL | 3.80T | 10 | $0.30 to $0.90 | 1.05M | 2026-10-15 |
| Nemotron 3 Ultra | 3.03T | 3 | $1.76 to $3.13 | 262K | 2026-06-04 |
| Nemotron 3.5 Lightning | 1.19T | 5 | $0.12 to $0.20 | 262K | 2026-08-11 |
| Nemotron 3 Nano Omni (free) | 120B | 0 | - | 256K | 2026-04-28 |
| Nemotron 3 Super | 13B | 3 | $0.40 to $2.40 | 1.05M | 2026-03-11 |
| Nemotron 3 Nano 30B A3B | 7B | 1 | $0.24 | 262K | 2025-12-14 |
| Nemotron 3.5 Content Safety | 64M | 1 | $0.20 | 131K | 2026-06-04 |
Finance your GPUs
Serving Nemotron on your own GPUs? Tell us the cluster, the models you serve and the contracts behind them, and we will share it with partner funders that fit.
Prefer email? hello@amcompute.com