Nemotron 3.5 Lightning API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of Nemotron 3.5 Lightning for your use-case.
Lowest Time per Task
Time per Intelligence Index task (minutes)
Total 5 providers
Lowest Cost per Task
Cost per Intelligence Index task
Total 5 providers
Nemotron 3.5 Lightning is available through 5 API providers, each offering different performance characteristics and pricing. Below is a comparison of the key metrics across providers.
- For time per task, PrimaLabs (NVFP4) (2.8 min), DeepInfra (NVFP4) (3.3 min), and CoreWeave (BF16) (4.0 min) finish an Intelligence Index task fastest.
- For cost per task, CoreWeave (BF16) ($0.24), Crusoe (BF16) ($0.26), and DeepInfra (NVFP4) ($0.31) are the cheapest per Intelligence Index task. Cost varies up to 2.3x across providers.
Pricing
Cost per Intelligence Index Task: Nemotron 3.5 Lightning
Pricing: Cache Hit, Input, and Output: Nemotron 3.5 Lightning
Output Speed vs. Cost per Task: Nemotron 3.5 Lightning
Speed
Average of the single-request and multi-turn agentic workloads measured
Time per Intelligence Index Task: Nemotron 3.5 Lightning
Latency
Average of the single-request and multi-turn agentic workloads measured
Time to First Answer Token: Nemotron 3.5 Lightning
Reliability
Share of requests the endpoint failed, across all the single-request and multi-turn agentic workloads measured
Endpoint Error Rate: Nemotron 3.5 Lightning
Caching
Cache Hit Rate: Nemotron 3.5 Lightning
Cost per Task vs. Cache Hit Rate: Nemotron 3.5 Lightning
Key Comparison Metrics & API Features
Frequently Asked Questions
Common questions about Nemotron 3.5 Lightning providers
Nemotron 3.5 Lightning is available through 5 API providers: DeepInfra (NVFP4), Nebius (BF16), CoreWeave (BF16), PrimaLabs (NVFP4), and Crusoe (BF16). Each provider offers different performance characteristics and pricing.
Nemotron 3.5 Lightning is currently available through 5 API providers that we benchmark and track.
By time per task, the fastest providers for Nemotron 3.5 Lightning are PrimaLabs (NVFP4) (2.8 min), DeepInfra (NVFP4) (3.3 min), and CoreWeave (BF16) (4.0 min). Time per task divides the model's output tokens per Intelligence Index task by each provider's end-to-end output speed.
The providers with the lowest time to first answer token for Nemotron 3.5 Lightning are PrimaLabs (NVFP4) (5.79s), DeepInfra (NVFP4) (9.84s), and CoreWeave (BF16) (10.49s). Lower latency means faster initial response time.
The most affordable providers for Nemotron 3.5 Lightning by blended price are Crusoe (BF16) ($0.051 per 1M tokens), CoreWeave (BF16) ($0.062 per 1M tokens), and DeepInfra (NVFP4) ($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 Nemotron 3.5 Lightning are Crusoe (BF16) ($0.05 per 1M input tokens), DeepInfra (NVFP4) ($0.06 per 1M input tokens), and Nebius (BF16) ($0.06 per 1M input tokens).
The providers with the lowest output token pricing for Nemotron 3.5 Lightning are DeepInfra (NVFP4) ($0.16 per 1M output tokens), CoreWeave (BF16) ($0.20 per 1M output tokens), and Crusoe (BF16) ($0.20 per 1M output tokens).
Prices for Nemotron 3.5 Lightning vary up to 2.0x across providers. The most affordable is Crusoe (BF16) at $0.051 per 1M tokens, while PrimaLabs (NVFP4) charges $0.10 per 1M tokens.
Time per task for Nemotron 3.5 Lightning varies significantly across providers. PrimaLabs (NVFP4) is the quickest at 2.8 minutes per Intelligence Index task, 2.1x quicker than Nebius (BF16) at 5.9 minutes.
All 5 providers of Nemotron 3.5 Lightning support JSON mode for structured output.
4 of 5 providers support function calling for Nemotron 3.5 Lightning: DeepInfra (NVFP4), Nebius (BF16), CoreWeave (BF16), and PrimaLabs (NVFP4).
For Nemotron 3.5 Lightning, PrimaLabs (NVFP4) offers the best performance with the lowest time per task and lowest latency. For cost optimization, Crusoe (BF16) provides the most competitive pricing.
When choosing a provider for Nemotron 3.5 Lightning, consider: time per task (how quickly the provider completes an Intelligence Index task), 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 Nemotron 3.5 Lightning's intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview