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 Results for other workloads are historical and no longer updated.

Kimi has launched a newer model, Kimi K2.6 (Non-reasoning). We suggest considering it instead.

For more information, see comparison of Kimi K2.6 (Non-reasoning) to other models and API provider benchmarks for Kimi K2.6 (Non-reasoning).

Kimi K2.5

Open weights model

Released 2026年1月

Kimi K2.5 (Non-reasoning) 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 Kimi K2.5 (Non-reasoning) 的最佳 API 服务商。

速度最快

#1
AzureAzure
170.9 t/s
#2
NovitaNovita
40.3 t/s
#3
GMIGMI
40.2 t/s

输出速度

共 3 家服务商

延迟最低

#1
AzureAzure
1.36 s
#2
NovitaNovita
1.97 s
#3
GMIGMI
2.80 s

首 Token 延迟

共 3 家服务商

价格最低

#1
NovitaNovita
$0.49
#2
GMIGMI
$0.84
#3
AzureAzure
$0.84

每 100 万 token 的混合价格

共 3 家服务商

可通过 3 家 API 服务商使用 Kimi K2.5 (Non-reasoning),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 Azure(170.9 t/s)、Novita(40.3 t/s)和GMI(40.2 t/s)。
  • 延迟方面,Azure(1.36 秒)、Novita(1.97 秒)和GMI(2.80 秒) 的首 Token 延迟最低。
  • 价格方面,Novita(0.49)、GMI(0.84)和Azure(0.84) 每 100 万 token 的混合价格最低。
  • Azure 兼具最高速度和最低延迟,性能表现最佳。若要优化成本,Novita 的价格最具竞争力。

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: Kimi K2.5 (Non-reasoning)

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

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

Novita 标志Novita
262k
开放
--
40
1.97
14.34
--
GMI 标志GMI
262k
开放
--
40
2.80
15.25
--
Microsoft Azure 标志Microsoft Azure
262k
开放
--
176
1.34
4.19
--

常见问题

关于 Kimi K2.5 (Non-reasoning) 服务商的常见问题