NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) for your use-case.
Fastest
Novita (FP4)
Crusoe (FP8)
DeepInfraOutput speed
Total 3 providers
Lowest Latency
Crusoe (FP8)DeepInfraNovita (FP4)Time to first token
Total 3 providers
Lowest Price
Crusoe (FP8)Novita (FP4)DeepInfraBlended price (per 1M tokens)
Total 3 providers
Nemotron 3 Nano (Non-reasoning) is available through 3 API providers, each offering different performance characteristics and pricing. Below is a comparison of the key metrics across providers.
- For output speed, the top providers are Novita (FP4) (310.2 t/s), Crusoe (FP8) (276.7 t/s), and DeepInfra (102.6 t/s).
- For latency, Crusoe (FP8) (0.37s), DeepInfra (0.70s), and Novita (FP4) (1.12s) offer the lowest time to first token.
- For pricing, Crusoe (FP8) (0.05), Novita (FP4) (0.07), and DeepInfra (0.07) offer the lowest blended prices per 1M tokens.
- Crusoe (FP8) offers both the lowest latency and best pricing, making it attractive for cost-conscious applications. Novita (FP4) is the fastest option for throughput-intensive workloads.
Highlights
Update: Default performance benchmarking workload has updated to 10k input tokens to better reflect production use cases. You can still select different workloads above.
Pricing
Pricing: Cache Hit, Input, and Output
Pricing: Blended Price
Pricing: Cache Discount
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed: NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)
Latency vs. Output Speed
Latency
Measured by Time (seconds) to First Token
Time to First Token
End-to-End Response Time
Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time
Key Comparison Metrics & API Features
Frequently Asked Questions
Common questions about NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) providers
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) is available through 3 API providers: Novita (FP4), DeepInfra, and Crusoe (FP8). Each provider offers different performance characteristics and pricing.
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) is currently available through 3 API providers that we benchmark and track.
The fastest providers for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) by output speed are Novita (FP4) (310.2 t/s), Crusoe (FP8) (276.7 t/s), and DeepInfra (102.6 t/s). Output speed measures how quickly tokens are generated after the model starts responding.
The providers with the lowest time to first token for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) are Crusoe (FP8) (0.37s), DeepInfra (0.70s), and Novita (FP4) (1.12s). Lower latency means faster initial response time.
The most affordable providers for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) by blended price are Crusoe (FP8) ($0.05 per 1M tokens), Novita (FP4) ($0.07 per 1M tokens), and DeepInfra ($0.07 per 1M tokens). Blended price uses a 7:2:1 cache hit/input/output token ratio.
The providers with the lowest input token pricing for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) are Novita (FP4) ($0.05 per 1M input tokens), DeepInfra ($0.05 per 1M input tokens), and Crusoe (FP8) ($0.05 per 1M input tokens).
The providers with the lowest output token pricing for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) are Novita (FP4) ($0.20 per 1M output tokens), DeepInfra ($0.20 per 1M output tokens), and Crusoe (FP8) ($0.20 per 1M output tokens).
Prices for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) vary up to 1.3x across providers. The most affordable is Crusoe (FP8) at $0.05 per 1M tokens, while Novita (FP4) charges $0.07 per 1M tokens.
Output speed for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) varies significantly across providers. Novita (FP4) is the fastest at 310.2 t/s, which is 3.0x faster than DeepInfra at 102.6 t/s.
All 3 providers of NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) support JSON mode for structured output.
All 3 providers of NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) support function calling (tool use).
The best provider for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) depends on your priorities: Novita (FP4) offers the highest output speed, Crusoe (FP8) has the lowest latency, and Crusoe (FP8) provides the most competitive pricing.
When choosing a provider for NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning), consider: output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and API features like JSON mode or function calling.
Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.
For information about NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)'s intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview