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 Results for other workloads are historical and no longer updated.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 Coder Next. We suggest considering it instead.

For more information, see comparison of Qwen3 Coder Next to other models and API provider benchmarks for Qwen3 Coder Next.

Open weights model

Released 2025年7月

Qwen3 Coder 480B A35B Instruct API 服务商基准测试与分析

模型比较

本分析旨在帮助你根据使用场景,选择 Qwen3 Coder 480B A35B Instruct 的最佳 API 服务商。

速度最快

#1
Google VertexGoogle Vertex
160.0 t/s
#2
AmazonAmazon
103.4 t/s
#3
DeepInfra (Turbo, FP4)DeepInfra (Turbo, FP4)
76.3 t/s
#4
Alibaba CloudAlibaba Cloud
52.7 t/s
#5
NovitaNovita
52.3 t/s

输出速度

共 5 家服务商

延迟最低

#1
DeepInfra (Turbo, FP4)DeepInfra (Turbo, FP4)
0.65 s
#2
Google VertexGoogle Vertex
0.78 s
#3
AmazonAmazon
1.21 s
#4
NovitaNovita
2.14 s
#5
Alibaba CloudAlibaba Cloud
3.09 s

首 Token 延迟

共 5 家服务商

价格最低

#1
DeepInfra (Turbo, FP4)DeepInfra (Turbo, FP4)
$0.23
#2
Google VertexGoogle Vertex
$0.38
#3
NovitaNovita
$0.50
#4
AmazonAmazon
$0.58
#5
Alibaba CloudAlibaba Cloud
$2.10

每 100 万 token 的混合价格

共 5 家服务商

可通过 5 家 API 服务商使用 Qwen3 Coder 480B,每家服务商的性能特征和价格各不相同。下方对比了各服务商的关键指标。

  • 输出速度最快的服务商是 Google Vertex(160.0 t/s)、Amazon(103.4 t/s)和DeepInfra (Turbo, FP4)(76.3 t/s)。 各服务商之间的速度差异显著,最快与最慢相差 206%。
  • 延迟方面,DeepInfra (Turbo, FP4)(0.65 秒)、Google Vertex(0.78 秒)和Amazon(1.21 秒) 的首 Token 延迟最低。
  • 价格方面,DeepInfra (Turbo, FP4)(0.23)、Google Vertex(0.38)和Novita(0.50) 每 100 万 token 的混合价格最低。 各服务商价格最多相差 9.1 倍。
  • DeepInfra (Turbo, FP4) 同时拥有最低延迟和最佳价格,对注重成本的应用很有吸引力。对于吞吐量密集型工作负载,Google Vertex 速度最快。

亮点

Output tokens per second · Higher is better
Seconds · Lower is better
USD per 1M tokens (blended) · Lower is better

价格

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens) · Lower is better · 10,000 input tokens

Pricing: Blended Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended) · Lower is better

Pricing: Cache Discount

1 - (cache hit price / input price) · Higher is better

Output Speed vs. Price

Blended at 7:2:1 (cache-input-output) · Output speed: output tokens per second · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

速度

按输出速度(每秒 token 数)衡量

Output Speed: Qwen3 Coder 480B A35B Instruct

Output speed: output tokens per second · 10,000 input tokens

Latency vs. Output Speed

Latency: seconds to first token received · Output speed: output tokens per second · 10,000 input tokens
Most attractive quadrant
Pareto line

延迟

按首 Token 延迟(秒)衡量

首 Token 延迟

Seconds to first token received · Lower is better · 10,000 input tokens

端到端响应时间

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

端到端响应时间

Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better · 10,000 input tokens

关键比较指标与 API 功能

Google 标志Google
262k
开放
--
161
0.78
3.89
--
Alibaba Cloud 标志Alibaba Cloud
262k
开放
--
54
3.12
12.45
--
DeepInfra 标志DeepInfra
262k
开放
--
79
0.64
6.93
--
Amazon Bedrock 标志Amazon Bedrock
262k
开放
--
104
1.21
5.99
--
Novita 标志Novita
262k
开放
--
53
2.12
11.47
--

常见问题

关于 Qwen3 Coder 480B A35B Instruct 服务商的常见问题