This model is deprecated. We only continue performance benchmarking for the default 10k input token workload. Results for other workloads are historical and no longer updated.
Alibaba has launched a newer model, Qwen3 30B A3B 2507 (Non-reasoning). We suggest considering it instead.
For more information, see comparison of Qwen3 30B A3B 2507 (Non-reasoning) to other models and API provider benchmarks for Qwen3 30B A3B 2507 (Non-reasoning).
Qwen3 32B (Non-reasoning) API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of Qwen3 32B (Non-reasoning) for your use-case.
Fastest
Amazon
Alibaba Cloud
DeepInfra (FP8)Output speed
Total 4 providers
Lowest Latency
AmazonDeepInfra (FP8)Alibaba CloudTime to first token
Total 4 providers
Lowest Price
DeepInfra (FP8)AmazonAlibaba Cloud
SambaNovaBlended price (per 1M tokens)
Total 4 providers
Qwen3 32B (Non-reasoning) 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 Amazon (111.9 t/s), Alibaba Cloud (99.3 t/s), and DeepInfra (FP8) (31.7 t/s).
- For latency, Amazon (1.10s), DeepInfra (FP8) (1.54s), and Alibaba Cloud (2.46s) offer the lowest time to first token.
- For pricing, DeepInfra (FP8) (0.10), Amazon (0.20), and Alibaba Cloud (0.21) offer the lowest blended prices per 1M tokens. Prices vary up to 4.4x across providers.
- Amazon offers the best performance with both the highest speed and lowest latency. For cost optimization, DeepInfra (FP8) provides the most competitive pricing.
Highlights
Pricing
Pricing: Cache Hit, Input, and Output
Pricing: Blended Price
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed: Qwen3 32B (Non-reasoning)
Latency vs. Output Speed
Latency
Measured by Time (seconds) to First Token
Time to First 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 Qwen3 32B (Non-reasoning) providers
Qwen3 32B (Non-reasoning) is available through 4 API providers: DeepInfra (FP8), Alibaba Cloud, Amazon, and SambaNova. Each provider offers different performance characteristics and pricing.
Qwen3 32B (Non-reasoning) is currently available through 4 API providers that we benchmark and track.
The fastest providers for Qwen3 32B (Non-reasoning) by output speed are Amazon (111.9 t/s), Alibaba Cloud (99.3 t/s), and DeepInfra (FP8) (31.7 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 Qwen3 32B (Non-reasoning) are Amazon (1.10s), DeepInfra (FP8) (1.54s), and Alibaba Cloud (2.46s). Lower latency means faster initial response time.
The most affordable providers for Qwen3 32B (Non-reasoning) by blended price are DeepInfra (FP8) ($0.10 per 1M tokens), Amazon ($0.20 per 1M tokens), and Alibaba Cloud ($0.21 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 Qwen3 32B (Non-reasoning) are DeepInfra (FP8) ($0.08 per 1M input tokens), Amazon ($0.15 per 1M input tokens), and Alibaba Cloud ($0.16 per 1M input tokens).
The providers with the lowest output token pricing for Qwen3 32B (Non-reasoning) are DeepInfra (FP8) ($0.28 per 1M output tokens), Amazon ($0.60 per 1M output tokens), and Alibaba Cloud ($0.64 per 1M output tokens).
Prices for Qwen3 32B (Non-reasoning) vary up to 4.4x across providers. The most affordable is DeepInfra (FP8) at $0.10 per 1M tokens, while SambaNova charges $0.44 per 1M tokens.
Output speed for Qwen3 32B (Non-reasoning) varies significantly across providers. Amazon is the fastest at 111.9 t/s, which is 3.5x faster than DeepInfra (FP8) at 31.7 t/s.
3 of 4 providers support JSON mode for Qwen3 32B (Non-reasoning): DeepInfra (FP8), Alibaba Cloud, and SambaNova.
All 4 providers of Qwen3 32B (Non-reasoning) support function calling (tool use).
For Qwen3 32B (Non-reasoning), Amazon offers the best performance with highest speed and lowest latency. For cost optimization, DeepInfra (FP8) provides the most competitive pricing.
When choosing a provider for Qwen3 32B (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 Qwen3 32B (Non-reasoning)'s intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview