LogoLMSpeed
  • Home
  • Free
  • Providers
  • Docs
LogoLMSpeed
LogoLMSpeed

The best API speed test tool

GitHubGitHubTwitterX (Twitter)Email
Product
  • Features
  • Pricing
  • FAQ
Leaderboard
  • Overview
  • Speed Ranking
  • Latency Ranking
  • Health Ranking
Models
  • All Models
  • GPT
  • Claude
  • Gemini
  • DeepSeek
  • Llama
  • Qwen
Free Models
  • All Free Models
  • Free GPT
  • Free Claude
  • Free Gemini
  • Free DeepSeek
  • Free Llama
  • Free Qwen
Resources
  • Speed Test
  • Provider Directory
  • Documentation
Legal
  • Cookie Policy
  • Privacy Policy
  • Terms of Service
© 2026 LMSpeed All Rights Reserved.Made by Nexmoe with ❤️
首页交流 QQ 群:1034193296,欢迎中转站站长加入讨论 AI 最热话题、newapi、openclaw 等,获取最新测速动态与反馈支持。
SiliconFlow logo

SiliconFlow

Provides cost-effective generative AI cloud services based on open-source models for text, image, video, and audio generation.

Categories

Country中国官方 API
ChatGLMGLM-Z1QwenDeepSeek R1 Distill Qwen 1 5bQwenQwen3 Omni InstructQwenQwen3 VL InstructQwenQwen3 5GeminiRing Flash 2.0QwenQwen3 Next InstructGeminiStep 3.5 FlashQwenQwen2 InstructQwenQwen3ChatGLMGLM-4ChatGLMGLM-4.7ChatGLMGLM-4.5VQwenDeepSeek R1 Distill QwenMinimaxMiniMax-M2.5QwenQwen2.5 Coder InstructChatGLMGLM-4.1v ThinkingQwenQwen2.5 InstructHunyuanHunyuan MtChatGLMGLM-4.6ChatGLMGLM-4.5 AirQwenDeepSeek R1 0528 Qwen3ChatGLMGLM-5MoonshotAIKimi K2.5QwenQwen3 Coder InstructDeepSeekDeepSeek V3.1 TerminusDeepSeekDeepSeek V3.1DeepSeekDeepSeek V3.2QwenQwen2.5 VL InstructMetaAIDeepSeek R1 Distill LlamaDeepSeekDeepSeek V3MoonshotAIKimi K2 InstructQwenQwen3 InstructDeepSeekDeepSeek R1DeepSeekDeepSeek V2.5

SiliconFlow offers 67 LLM API models.

Speed benchmark average: 46 tok/s.

SiliconFlow interface preview
Avg Speed45.81 tok/s
Latency13.05 s
Total Tests1197
Models67
Updated4/17/2026
Created At8/13/2025
Website

API Endpoints

  • Historical / Unverified
    https://account.siliconflow.cn
  • Historical / Unverified
    https://cloud.siliconflow.cn
  • Historical / Unverified
    https://cloud.siliconflow.com
  • Historical / Unverified
    https://api.siliconflow.cn
  • Historical / Unverified
    https://api.siliconflow.com

Supported Models

ModelSpeedLatencyTests
PaddlePaddle/PaddleOCR-VL-1.5
279.58 tok/s
4.15s
3
ChatGLMTHUDM/GLM-Z1-9B-0414
176.03 tok/s
13.40s
24
Qwendeepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
142.95 tok/s
4.47s
5
QwenQwen/Qwen3-Omni-30B-A3B-Instruct
128.77 tok/s
0.47s
5
QwenQwen/Qwen3-VL-8B-Instruct
106.50 tok/s
0.92s
10
QwenQwen/Qwen3.5-4B
91.28 tok/s
23.54s
5
GeminiinclusionAI/Ring-flash-2.0
89.88 tok/s
6.43s
5
QwenQwen/Qwen3-Next-80B-A3B-Instruct
86.82 tok/s
0.62s
10
Geministepfun-ai/Step-3.5-Flash
84.22 tok/s
3.37s
5
QwenQwen/Qwen2-7B-Instruct
81.93 tok/s
0.57s
25
QwenQwen/Qwen3-14B
78.64 tok/s
9.81s
5
ChatGLMPro/THUDM/glm-4-9b-chat
76.25 tok/s
0.63s
10
ChatGLMTHUDM/glm-4-9b-chat
75.38 tok/s
0.59s
15
ChatGLMzai-org/GLM-4.7
74.72 tok/s
16.10s
5
ChatGLMzai-org/GLM-4.5V
73.00 tok/s
6.15s
10
Qwendeepseek-ai/DeepSeek-R1-Distill-Qwen-7B
72.09 tok/s
8.93s
30
ChatGLMTHUDM/GLM-4-9B-0414
71.99 tok/s
0.95s
15
MinimaxPro/MiniMaxAI/MiniMax-M2.5
71.37 tok/s
9.63s
25
QwenPro/Qwen/Qwen2-7B-Instruct
71.29 tok/s
0.56s
5
QwenQwen/QwQ-32B-Preview
69.75 tok/s
0.62s
5
Showing 20 of 67 models.

Leaderboard Rankings

Speed
83.1 tokens/s#29/100
Latency
0.24 s#2/100
OverviewPerformance67PricingTests1197HealthEmbed