DeepSeek V3.2 (Non-reasoning) API Provider Benchmarking & Analysis
Analysis of API providers for DeepSeek V3.2 (Non-reasoning) across performance metrics including latency (time to first token), output speed (output tokens per second), price and others. API providers benchmarked include Amazon Bedrock, GMI, SiliconFlow (FP8), DeepSeek, Google Vertex, Fireworks, Novita, DeepInfra, Eigen AI, Microsoft Azure, Nebius, FriendliAI.
Fastest
Fireworks
FriendliAI
Eigen AI
Nebius
AmazonOutput speed
Total 12 providers
Lowest Latency
Fireworks
DeepInfraFriendliAIDeepSeekEigen AITime to first token
Total 12 providers
Lowest Price
GMIDeepInfra
SiliconFlow (FP8)DeepSeek
NovitaBlended price (per 1M tokens)
Total 12 providers
DeepSeek V3.2 is available through 12 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 Fireworks (203.0 t/s), FriendliAI (104.2 t/s), Eigen AI (79.6 t/s). Speed varies significantly across providers, with a 211% difference between the fastest and slowest.
- For latency, Fireworks (0.80s), DeepInfra (1.07s), FriendliAI (1.07s) offer the lowest time to first token.
- For pricing, GMI (0.23), DeepInfra (0.29), SiliconFlow (FP8) (0.31) offer the lowest blended prices per 1M tokens.
- Fireworks offers the best performance with both the highest speed and lowest latency. For cost optimization, GMI provides the most competitive pricing.
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: Input and Output Prices: DeepSeek V3.2 Providers
Speed vs. Price: DeepSeek V3.2 Providers
Speed
Measured by Output Speed (tokens per second)
Output Speed: DeepSeek V3.2 Providers
Latency vs. Output Speed: DeepSeek V3.2 Providers
Latency
Measured by Time (seconds) to First Token
Time to First Token: DeepSeek V3.2 Providers
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: DeepSeek V3.2 Providers
API Features
Function (Tool) Calling & JSON Mode: DeepSeek V3.2 Providers
SiliconFlow (FP8)
Google Vertex
DeepInfra
AzureContext Window: DeepSeek V3.2 Providers
Summary Table of Key Comparison Metrics
Frequently Asked Questions
Common questions about DeepSeek V3.2 (Non-reasoning) providers
DeepSeek V3.2 (Non-reasoning) is available through 12 API providers: Amazon, GMI, SiliconFlow (FP8), DeepSeek, Google Vertex, Fireworks, Novita, DeepInfra, Eigen AI, Azure, Nebius, and FriendliAI. Each provider offers different performance characteristics and pricing.
DeepSeek V3.2 (Non-reasoning) is currently available through 12 API providers that we benchmark and track.
The fastest providers for DeepSeek V3.2 (Non-reasoning) by output speed are Fireworks (203.0 t/s), FriendliAI (104.2 t/s), and Eigen AI (79.6 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 DeepSeek V3.2 (Non-reasoning) are Fireworks (0.80s), DeepInfra (1.07s), and FriendliAI (1.07s). Lower latency means faster initial response time.
The most affordable providers for DeepSeek V3.2 (Non-reasoning) by blended price are GMI ($0.23 per 1M tokens), DeepInfra ($0.29 per 1M tokens), and SiliconFlow (FP8) ($0.31 per 1M tokens). Blended price uses a 3:1 input to output token ratio.
The providers with the lowest input token pricing for DeepSeek V3.2 (Non-reasoning) are GMI ($0.20 per 1M input tokens), DeepInfra ($0.26 per 1M input tokens), and SiliconFlow (FP8) ($0.27 per 1M input tokens).
The providers with the lowest output token pricing for DeepSeek V3.2 (Non-reasoning) are GMI ($0.32 per 1M output tokens), DeepInfra ($0.38 per 1M output tokens), and SiliconFlow (FP8) ($0.42 per 1M output tokens).
9 of 12 providers support JSON mode for DeepSeek V3.2 (Non-reasoning): SiliconFlow (FP8), DeepSeek, Google Vertex, Novita, DeepInfra, Eigen AI, Azure, Nebius, and FriendliAI.
8 of 12 providers support function calling for DeepSeek V3.2 (Non-reasoning): Amazon, SiliconFlow (FP8), DeepSeek, Google Vertex, DeepInfra, Eigen AI, Azure, and Nebius.
When choosing a provider for DeepSeek V3.2 (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 DeepSeek V3.2 (Non-reasoning)'s intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview