Beam 1.0 API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of Beam 1.0 for your use-case.
Lowest Time per Task
Baseten (NVFP4)
Nebius (NVFP4)
Together AI (NVFP4)
CoreWeave (NVFP4)
DigitalOceanTime per Intelligence Index task (minutes)
Total 5 providers
Lowest Cost per Task
DigitalOceanTogether AI (NVFP4)Baseten (NVFP4)Nebius (NVFP4)CoreWeave (NVFP4)Cost per Intelligence Index task
Total 5 providers
Beam 1.0 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, Baseten (NVFP4) (1.2 min), Nebius (NVFP4) (1.8 min), and Together AI (NVFP4) (2.3 min) finish an Intelligence Index task fastest.
- For cost per task, DigitalOcean ($0.28), Together AI (NVFP4) ($0.28), and Baseten (NVFP4) ($0.45) are the cheapest per Intelligence Index task. Cost varies up to 2.5x across providers.
Pricing
Cost per Intelligence Index Task: Beam 1.0
Pricing: Cache Hit, Input, and Output: Beam 1.0
Output Speed vs. Cost per Task: Beam 1.0
Speed
Average of the single-request and multi-turn agentic workloads measured
Time per Intelligence Index Task: Beam 1.0
Latency
Average of the single-request and multi-turn agentic workloads measured
Time to First Answer Token: Beam 1.0
Reliability
Share of requests the endpoint failed, across all the single-request and multi-turn agentic workloads measured
Endpoint Error Rate: Beam 1.0
Caching
Cache Hit Rate: Beam 1.0
Cost per Task vs. Cache Hit Rate: Beam 1.0
Key Comparison Metrics & API Features
Frequently Asked Questions
Common questions about Beam 1.0 providers
Beam 1.0 is available through 5 API providers: Baseten (NVFP4), CoreWeave (NVFP4), Together AI (NVFP4), DigitalOcean, and Nebius (NVFP4). Each provider offers different performance characteristics and pricing.
Beam 1.0 is currently available through 5 API providers that we benchmark and track.
By time per task, the fastest providers for Beam 1.0 are Baseten (NVFP4) (1.2 min), Nebius (NVFP4) (1.8 min), and Together AI (NVFP4) (2.3 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 Beam 1.0 are Baseten (NVFP4) (4.38s), Nebius (NVFP4) (6.31s), and CoreWeave (NVFP4) (7.17s). Lower latency means faster initial response time.
The most affordable providers for Beam 1.0 by blended price are Nebius (NVFP4) ($0.18 per 1M tokens), Baseten (NVFP4) ($0.23 per 1M tokens), and Together AI (NVFP4) ($0.23 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 Beam 1.0 are Nebius (NVFP4) ($0.25 per 1M input tokens), Baseten (NVFP4) ($0.40 per 1M input tokens), and Together AI (NVFP4) ($0.40 per 1M input tokens).
The providers with the lowest output token pricing for Beam 1.0 are Nebius (NVFP4) ($1.00 per 1M output tokens), Baseten (NVFP4) ($1.20 per 1M output tokens), and Together AI (NVFP4) ($1.20 per 1M output tokens).
Prices for Beam 1.0 vary up to 3.1x across providers. The most affordable is Nebius (NVFP4) at $0.18 per 1M tokens, while CoreWeave (NVFP4) charges $0.57 per 1M tokens.
Time per task for Beam 1.0 varies significantly across providers. Baseten (NVFP4) is the quickest at 1.2 minutes per Intelligence Index task, 2.3x quicker than DigitalOcean at 2.7 minutes.
All 5 providers of Beam 1.0 support JSON mode for structured output.
All 5 providers of Beam 1.0 support function calling (tool use).
For Beam 1.0, Baseten (NVFP4) offers the best performance with the lowest time per task and lowest latency. For cost optimization, Nebius (NVFP4) provides the most competitive pricing.
When choosing a provider for Beam 1.0, 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 Beam 1.0's intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview