Hy3 API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of Hy3 for your use-case.
Fastest
GMI
SiliconFlow (FP8)
Novita
DeepInfra (FP8)Output speed
Total 4 providers
Lowest Latency
SiliconFlow (FP8)GMINovitaDeepInfra (FP8)Time to first answer token
Total 4 providers
Lowest Price
SiliconFlow (FP8)NovitaDeepInfra (FP8)GMIBlended price (per 1M tokens)
Total 4 providers
Hy3 is available through 4 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 GMI (96.5 t/s), SiliconFlow (FP8) (96.2 t/s), and Novita (71.4 t/s). Speed varies significantly across providers, with a 63% difference between the fastest and slowest.
- For latency, SiliconFlow (FP8) (23.71s), GMI (23.77s), and Novita (30.23s) offer the lowest time to first answer token.
- For pricing, SiliconFlow (FP8) (0.10), Novita (0.11), and DeepInfra (FP8) (0.11) offer the lowest blended prices per 1M tokens.
- SiliconFlow (FP8) offers both the lowest latency and best pricing, making it attractive for cost-conscious applications. GMI 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: Hy3
Latency vs. Output Speed
Latency
Measured by Time (seconds) to First Token
Time to First Answer 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 Hy3 providers
Hy3 is available through 4 API providers: SiliconFlow (FP8), Novita, DeepInfra (FP8), and GMI. Each provider offers different performance characteristics and pricing.
Hy3 is currently available through 4 API providers that we benchmark and track.
The fastest providers for Hy3 by output speed are GMI (96.5 t/s), SiliconFlow (FP8) (96.2 t/s), and Novita (71.4 t/s). Output speed measures how quickly tokens are generated after the model starts responding.
The providers with the lowest time to first answer token for Hy3 are SiliconFlow (FP8) (23.71s), GMI (23.77s), and Novita (30.23s). Lower latency means faster initial response time.
The most affordable providers for Hy3 by blended price are SiliconFlow (FP8) ($0.10 per 1M tokens), Novita ($0.11 per 1M tokens), and DeepInfra (FP8) ($0.11 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 Hy3 are SiliconFlow (FP8) ($0.13 per 1M input tokens), Novita ($0.14 per 1M input tokens), and DeepInfra (FP8) ($0.14 per 1M input tokens).
The providers with the lowest output token pricing for Hy3 are SiliconFlow (FP8) ($0.53 per 1M output tokens), Novita ($0.58 per 1M output tokens), and DeepInfra (FP8) ($0.58 per 1M output tokens).
Output speed for Hy3 varies significantly across providers. GMI is the fastest at 96.5 t/s, which is 1.6x faster than DeepInfra (FP8) at 59.0 t/s.
2 of 4 providers support JSON mode for Hy3: DeepInfra (FP8) and GMI.
All 4 providers of Hy3 support function calling (tool use).
The best provider for Hy3 depends on your priorities: GMI offers the highest output speed, SiliconFlow (FP8) has the lowest latency, and SiliconFlow (FP8) provides the most competitive pricing.
When choosing a provider for Hy3, 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 Hy3's intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview