This model is deprecated. We still benchmark its performance, but less often: every 18 hours instead of every 6.
NVIDIA has launched a newer release, Nemotron 3.5 Super VL. We suggest considering it instead.
Nemotron 3 Super 120B A12B (Reasoning) API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of Nemotron 3 Super 120B A12B (Reasoning) for your use-case.
Lowest Time per Task
Time per Intelligence Index task (minutes)
Total 3 providers
Lowest Cost per Task
Cost per Intelligence Index task
Total 3 providers
Nemotron 3 Super is available through 3 API providers, each offering different performance characteristics and pricing. Below is a comparison of the key metrics across providers.
Pricing
Cost per Intelligence Index Task: Nemotron 3 Super 120B A12B (Reasoning)
Pricing: Cache Hit, Input, and Output: Nemotron 3 Super 120B A12B (Reasoning)
Output Speed vs. Cost per Task: Nemotron 3 Super 120B A12B (Reasoning)
Speed
Average of the single-request and multi-turn agentic workloads measured
Output Speed: Nemotron 3 Super 120B A12B (Reasoning)
Latency
Average of the single-request and multi-turn agentic workloads measured
Time to First Answer Token: Nemotron 3 Super 120B A12B (Reasoning)
Reliability
Share of requests the endpoint failed, across all the single-request and multi-turn agentic workloads measured
Endpoint Error Rate: Nemotron 3 Super 120B A12B (Reasoning)
Caching
Cache Hit Rate: Nemotron 3 Super 120B A12B (Reasoning)
Cost per Task vs. Cache Hit Rate: Nemotron 3 Super 120B A12B (Reasoning)
Key Comparison Metrics & API Features
Frequently Asked Questions
Common questions about Nemotron 3 Super 120B A12B (Reasoning) providers
Nemotron 3 Super 120B A12B (Reasoning) is available through 3 API providers: Crusoe (FP8), Nebius, and DeepInfra. Each provider offers different performance characteristics and pricing.
Nemotron 3 Super 120B A12B (Reasoning) is currently available through 3 API providers that we benchmark and track.
The providers with the lowest time to first answer token for Nemotron 3 Super 120B A12B (Reasoning) are DeepInfra (9.15s), Nebius (11.58s), and Crusoe (FP8) (51.23s). Lower latency means faster initial response time.
The most affordable providers for Nemotron 3 Super 120B A12B (Reasoning) by blended price are DeepInfra ($0.11 per 1M tokens), Nebius ($0.19 per 1M tokens), and Crusoe (FP8) ($0.40 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 Super 120B A12B (Reasoning) are DeepInfra ($0.075 per 1M input tokens), Crusoe (FP8) ($0.30 per 1M input tokens), and Nebius ($0.30 per 1M input tokens).
The providers with the lowest output token pricing for Nemotron 3 Super 120B A12B (Reasoning) are DeepInfra ($0.40 per 1M output tokens), Nebius ($0.90 per 1M output tokens), and Crusoe (FP8) ($2.40 per 1M output tokens).
Prices for Nemotron 3 Super 120B A12B (Reasoning) vary up to 3.8x across providers. The most affordable is DeepInfra at $0.11 per 1M tokens, while Crusoe (FP8) charges $0.40 per 1M tokens.
All 3 providers of Nemotron 3 Super 120B A12B (Reasoning) support JSON mode for structured output.
All 3 providers of Nemotron 3 Super 120B A12B (Reasoning) support function calling (tool use).
When choosing a provider for Nemotron 3 Super 120B A12B (Reasoning), 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 Super 120B A12B (Reasoning)'s intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview