Sponsored byFusecodeEnterprise coding API for Claude Code, Codex, and model workflows.
LogoLMSpeed
  • Free
  • Models
  • Providers
  • Docs
LogoLMSpeed
  1. Home
  2. Models
  3. Categories
  4. LLM
LogoLMSpeed

The best API speed test tool

GitHubGitHubTwitterX (Twitter)Email
Product
  • Features
  • Pricing
  • FAQ
Leaderboard
  • Overview
  • Speed Ranking
  • Latency Ranking
  • Health Ranking
  • Model Pricing
  • Model Speed
  • Reasoning
  • Coding
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
Tools
  • Speed Test
  • Provider Audit
Resources
  • Provider Directory
  • Documentation
  • GEO Checker
  • Viora
  • Botab
  • VidBee
Legal
  • Cookie Policy
  • Privacy Policy
  • Terms of Service
© 2026 LMSpeed All Rights Reserved.Made by Nexmoe with ❤️
All347LLM (Text-to-text)342Image Generation5
Newest342Most tested0Recently tested0A–Z0
Top for reasoning65.3
  • OpenAIGPT-5 Codex
  • MinimaxMiniMax M2.1
  • MoonshotAIKimi K2 Thinking
On this page
Top for coding66.3
  • OpenAIGPT-5.4
  • MoonshotAIKimi K2 Thinking
  • OpenAIGPT-5.1 Codex Mini
On this page
Lowest latency0.719s
  • ChatGLMGLM-4
  • DeepSeekDeepSeek VL2
  • Devstral 2
On this page
Highest throughput277 t/s
  • GeminiGemini 2.5 Flash Lite
  • Ministral 3
  • QwenQwen3 VL
On this page
342 Visible models
ModelContextInputOutputThroughputLatencyProvidersAgentsCodingReasoningKnowledgeMathMultilingualMultimodalInstruction following
OpenAIGPT-5.4 Pro1.1M————
+60
89.3—57.338.771.8—94—
OpenAIGPT-5.41.1M——51 t/s4.45s
+262
63.866.345.756.2——75.673.9
OpenAIGPT-5.3128K——59 t/s2.57s
+67
————————
QwenQwen3.5 Flash1M——115 t/s7.92s
+60
————————
OpenAIGPT-5.3 Codex400K——79 t/s4.08s
+229
71.460.255.654——89.375.4
GeminiGemini 3.1 Pro1.0M——92 t/s15.36s
+165
52.262.655.855.5——78.377.1
ClaudeClaude Sonnet 4.61M——43 t/s4.13s
+193
67.853.951.154.6——82.741.2
MinimaxMiniMax M2.5204.8K——57 t/s9.85s
+167
—28.752—————
ChatGLMGLM-5202.8K——46 t/s22.00s
+166
50.457.654.95092.169.110082.4
ClaudeClaude Opus 4.61M——51 t/s5.32s
+200
63.259.549.153.899.8—78.544.6
StepfunStep 3.5 Flash262.1K——104 t/s6.74s
+57
—38.552.6—————
MoonshotAIKimi K2.5262.1K——146 t/s11.91s
+174
48.85951.362.391.969.284.782.1
ChatGLMGLM-4.7 Flash202.8K——49 t/s23.19s
+44
—33.732.6—————
OpenAIGPT-5.2 Codex400K——54 t/s3.51s
+145
71.946.15255.4——76.377.6
MinimaxMiniMax M2.1204.8K——74 t/s4.19s
+95
—60.964.2—————
ChatGLMGLM-4.7202.8K——94 t/s18.83s
+135
50.449.543.552.995.7—10067.9
GeminiGemini 3 Flash1.0M——144 t/s8.07s
+165
49.746.346.651.7——89.355.1
MiMo-V2-Flash262.1K——107 t/s3.20s
+48
5342.252.337.794.1——39.9
ChatGLMGLM-4———63 t/s0.72s
+47
————————
DeepSeekDeepSeek VL2———126 t/s0.75s
+8
————————
MAI-DS-R1—————
+7
————————
Intern-S1—————
+3
————————
QwenQwen Image—————
+34
————————
DeepSeekDeepSeek Coder—————
+8
————————
QwenQwen3 Next—————
+7
————————
QwenQwen3 Embedding—————
+36
————————
MistralMistral Large 3———29 t/s3.00s
+5
34.13146.939.2——55.736.2
Ministral 3———208 t/s2.84s
+1
—29.343.7—————
Devstral 2———71 t/s1.34s
+3
—3946.4—————
QwenQwen3 VL———206 t/s6.63s
+16
————————
DeepSeekDeepSeek V3———49 t/s3.39s
+142
41.529.840.937.757—10060.5
OpenAIGPT-5.2400K——56 t/s3.43s
+223
61.156.549.558.3——84.175.4
OpenAIGPT-5.2 Pro400K————
+76
————————
ChatGLMGLM-4.6V131.1K——40 t/s27.37s
+50
—34.245.2—————
OpenAIGPT-5.1 Codex Max400K——56 t/s6.29s
+116
8332.436.551.5——72.570
DeepSeekDeepSeek V3.2131.1K——46 t/s5.36s
+164
41.847.746.545.6——10049
ClaudeClaude Opus 4.5200K——54 t/s3.12s
+145
57.361.455.36390.171.272.764
OpenAIGPT-5.1400K——142 t/s2.78s
+142
53.748.856.140.9——87.872.9
OpenAIGPT-5.1 Codex400K————
+123
66.342.653.751.5——86.370
OpenAIGPT-5.1 Codex Mini400K——136 t/s4.28s
+115
—63.160.1—————
MoonshotAIKimi K2 Thinking262.1K——51 t/s16.73s
+95
—63.963.6—————
Nova Premier V11M————
+2
————————
MinimaxMiniMax M2204.8K——48 t/s10.12s
+70
—59.457.4—————
Granite 4.0 H Micro131K————
+4
————————
AzurePhi 4 Mini Instruct131.1K————
+24
—11.727.9—36.3———
ClaudeClaude Haiku 4.5200K——102 t/s2.77s
+172
—52.949.6———100—
OpenAIO3 Deep Research200K————
+23
————————
OpenAIO4 Mini Deep Research200K————
+21
————————
OpenAIGPT-5 Pro400K————
+61
————————
ChatGLMGLM-4.6202.8K——53 t/s16.68s
+89
76.934.650.135.7———36.7
ClaudeClaude Sonnet 4.51M——41 t/s4.57s
+154
44.259.744.8—87—100—
GeminiGemini 2.5 Flash Lite1.0M——277 t/s1.41s
+106
—39.348.3—83.6———
QwenQwen3 Max262.1K——38 t/s7.37s
+83
5932.954.145.1——10044.1
QwenQwen3 Coder Plus1M——45 t/s1.35s
+64
————————
OpenAIGPT-5 Codex400K——154 t/s2.76s
+117
—62.565.3—————
DeepSeekDeepSeek V3.1 Terminus163.8K——58 t/s1.50s
+81
—42.555.7—————
MoonshotAIKimi K2131.1K——29 t/s2.21s
+79
61.135.154.340.883.2—10041.5
DeepSeekDeepSeek Chat V3.1163.8K——23 t/s2.02s
+5
————————
MistralMistral Medium 3.1131.1K————
+1
—37.243.8—————
ChatGLMGLM-4.5V65.5K——73 t/s6.15s
+32
—2745.3—————
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6