Same open weights, served by 2 providers. Weights: nvidia/Nemotron-3-Nano-30B-A3B.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
DeepInfra Deprecated, 2026-10-08 | nvidia/Nemotron-3-Nano-30B-A3B | flexFP40.8× standard | $0.04 | $0.16 | $0.02 | $0.07 | Verified 8h ago · Changed 10h ago | |
DeepInfra Deprecated, 2026-10-08 | nvidia/Nemotron-3-Nano-30B-A3B | standardFP4 | $0.05 | $0.20 | $0.025 | $0.088 | Verified 8h ago · Changed 10h ago | |
AWS Bedrock | nvidia.nemotron-nano-3-30b | standardQuantization not disclosed | $0.06 | $0.24 | — | $0.105 | Verified 10h ago · Changed 10h ago | |
DeepInfra Deprecated, 2026-10-08 | nvidia/Nemotron-3-Nano-30B-A3B | priorityFP41.5× standard | $0.075 | $0.30 | $0.037 | $0.131 | Verified 8h ago · Changed 10h ago |
Same weights don't always mean identical outputs: quantization, context limits and serving stacks can change quality, especially for tool calling. Prices are USD per 1M tokens; hyperscaler prices are for the reference region shown.
| DeepInfra fp4 | $9/mo |
| AWS Bedrock | $10.8/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.