Skip to content

编程AI性能追踪器

所有编程AI模型的每日排名变动、评分趋势和性能数据。评分每小时从实时模型数据更新。

模型总数
300
上升 (24小时)
11
下降 (24小时)
1
不变
288

按服务商平均评分 (前12)

LMMarketCap.com

评分分布

LMMarketCap.com

24小时最大变动

涨幅最大
Inkling (batch)thinkingmachines · #144
+1
GLM 5V TurboZhipu AI · #145
+1
o4 Mini HighOpenAI · #146
+1
MiniMax M2.1MiniMax · #147
+1
MiniMax M2MiniMax · #148
+1
跌幅最大
Inkling Smallthinkingmachines · #154
-10

每日性能

#模型提供商评分24小时7天状态14天趋势
1Claude Fable 5Anthropic97.100stable
2Claude Fable 5 (batch)Anthropic97.10+377preliminary
3Claude Opus 5 (Fast)Anthropic95.10+171fragile
4Claude Opus 5Anthropic95.10+171fragile
5Claude Opus 4.8 (Fast)Anthropic95.10-1stable
6Claude Opus 4.8Anthropic95.10-1stable
7Claude Opus 4.7 (Fast)Anthropic95.10-5stable
8Claude Opus 4.7Anthropic95.10-5stable
9Claude Opus 4.7 (batch)Anthropic95.10+370preliminary
10Claude Opus 4.8 (batch)Anthropic94.60+369preliminary
11GPT-5.5 ProOpenAI92.700stable
12GPT-5.5 Pro (batch)OpenAI92.70+367preliminary
13GPT-5.5OpenAI92.70-7fragile
14GPT-5.5 (batch)OpenAI92.70+365preliminary
15Gemini 3.1 Pro Preview Custom ToolsGoogle92.20-8fragile
16Gemini 3.1 Pro PreviewGoogle92.20-8fragile
17Gemini 3.1 Pro Preview (batch)Google92.20+362preliminary
18GPT-5.4 ProOpenAI91.90-9fragile
19GPT-5.4 Pro (batch)OpenAI91.90+360preliminary
20GPT-5.4OpenAI91.90-10fragile
21GPT-5.4 (batch)OpenAI91.90+358preliminary
22GPT-5.3 ChatOpenAI90.50+10fragile
23GPT-5.3-CodexOpenAI90.500stable
24GPT-5.2-CodexOpenAI90.50-12fragile
25GPT-5.2 ChatOpenAI90.50+40fragile
26GPT-5.2 ProOpenAI90.50-13fragile
27GPT-5.2 Pro (batch)OpenAI90.50+352preliminary
28GPT-5.2OpenAI90.50-14fragile
29GPT-5.2 (batch)OpenAI90.50+350preliminary
30Claude Opus 4.6Anthropic90.40-15fragile
31Claude Opus 4.6 (batch)Anthropic90.40+348preliminary
32GPT-5.6 Luna ProOpenAI89.00-16fragile
33GPT-5.6 Luna Pro (batch)OpenAI89.00+346preliminary
34GPT-5.6 LunaOpenAI89.00-17fragile
35GPT-5.6 Luna (batch)OpenAI89.00+344preliminary
36GPT-5.6 Terra ProOpenAI89.00-18fragile
37GPT-5.6 Terra Pro (batch)OpenAI89.00+342preliminary
38GPT-5.6 TerraOpenAI89.00-19fragile
39GPT-5.6 Terra (batch)OpenAI89.00+340preliminary
40GPT-5.6 Sol ProOpenAI89.00-20fragile
41GPT-5.6 Sol Pro (batch)OpenAI89.00+338preliminary
42GPT-5.6 SolOpenAI89.00-21fragile
43GPT-5.6 Sol (batch)OpenAI89.00+336preliminary
44Grok 4.5xAI88.80+335preliminary
45Grok 4.3xAI88.80+334preliminary
46Grok 4.20xAI88.80+333preliminary
47GPT-5.1-Codex-MaxOpenAI88.70-19fragile
48GPT-5.1OpenAI88.70-19fragile
49GPT-5.1-CodexOpenAI88.70-19fragile
50GPT-5 ProOpenAI88.70-26fragile
51GPT-5 Pro (batch)OpenAI88.70+328preliminary
52GPT-5 Codex (batch)OpenAI88.70+327preliminary
53GPT-5OpenAI88.70-28fragile
54GPT-5 (batch)OpenAI88.70+325preliminary
55Gemini 3 Flash PreviewGoogle88.40-29fragile
56Gemini 3 Flash Preview (batch)Google88.40+323preliminary
57Grok 4.20 Multi-AgentxAI87.90+322preliminary
58GPT-5.1 (batch)OpenAI87.80+321preliminary
59GPT-5.1-Codex-MiniOpenAI87.80-28fragile
60DeepSeek V4 ProDeepSeek86.70-27fragile
61o3 ProOpenAI86.70-27fragile
62o3 Pro (batch)OpenAI86.70+317preliminary
63o3OpenAI86.70-28fragile
64o3 (batch)OpenAI86.70+315preliminary
65Claude Sonnet 5Anthropic85.20+123fragile
66Claude Sonnet 4.6Anthropic85.20-30fragile
67Claude Sonnet 4.6 (batch)Anthropic85.20+312preliminary
68Claude Opus 4.5Anthropic85.10-31fragile
69Claude Opus 4.5 (batch)Anthropic85.10+310preliminary
70Gemini 2.5 ProGoogle83.50-32fragile
71Gemini 2.5 Pro (batch)Google83.50+308preliminary
72Gemini 2.5 Pro Preview 06-05Google83.50-33fragile
73Gemini 2.5 Pro Preview 05-06Google83.50-33fragile
74Claude Sonnet 4.5Anthropic82.40-33fragile
75Claude Sonnet 4.5 (batch)Anthropic82.40+304preliminary
76Claude Opus 4.1Anthropic82.10-2stable
77Claude Opus 4Anthropic82.10-35fragile
78o4 Mini High (batch)OpenAI81.40+301preliminary
79o4 MiniOpenAI81.40-36fragile
80o4 Mini (batch)OpenAI81.40+299preliminary
81DeepSeek V3.2DeepSeek81.30-37fragile
82Qwen3.8 MaxAlibaba80.60+297preliminary
83Gemma 4 31BGoogle80.50-36fragile
84Gemma 4 31B (free)Google80.50-36fragile
85Muse Spark 1.2meta80.20+294preliminary
86Muse Spark 1.1meta80.20-40fragile
87Gemini 3.6 FlashGoogle80.00-38fragile
88Gemini 3.6 Flash (batch)Google80.00+291preliminary
89Qwen3.5 397B A17BAlibaba79.40-39fragile
90R1 0528DeepSeek79.40-39fragile
91GPT-5.4 NanoOpenAI79.30-39fragile
92GPT-5.4 Nano (batch)OpenAI79.30+287preliminary
93GPT-5.4 MiniOpenAI79.30-40fragile
94GPT-5.4 Mini (batch)OpenAI79.30+285preliminary
95Gemini 3.1 Flash Lite PreviewGoogle79.30-41fragile
96Gemini 2.5 Flash LiteGoogle79.10-41fragile
97Gemini 2.5 Flash Lite (batch)Google79.10+282preliminary
98Gemini 2.5 FlashGoogle79.10-42fragile
99Gemini 2.5 Flash (batch)Google79.10+280preliminary
100Gemini 3.5 FlashGoogle79.00-43fragile
101Gemini 3.5 Flash (batch)Google79.00+278preliminary
102GLM 5.2Zhipu AI78.50-44fragile
103GLM 5.2 (batch)Zhipu AI78.30+276preliminary
104GLM 5.1Zhipu AI78.00-32fragile
105MiniMax M2.7MiniMax78.00-1stable
106GLM 5 TurboZhipu AI78.00-8fragile
107MiniMax M2.5MiniMax78.00-48fragile
108GLM 5Zhipu AI78.00-48fragile
109Qwen3.5-122B-A10BAlibaba77.70-46fragile
110Gemma 2 27BGoogle77.40-46fragile
111Qwen3.5-27BAlibaba77.00-45fragile
112Gemini 3.5 Flash LiteGoogle76.90-45fragile
113Gemini 3.5 Flash Lite (batch)Google76.90+266preliminary
114GPT-5 Mini (batch)OpenAI76.80+265preliminary
115GPT-5 Nano (batch)OpenAI76.80+264preliminary
116MiMo-V2.5-ProXiaomi76.20-47fragile
117Qwen3.5-35B-A3BAlibaba76.00-47fragile
118Qwen3.7 PlusAlibaba75.90-50fragile
119Kimi K2.6Moonshot AI75.80-48fragile
120o3 MiniOpenAI75.30-47fragile
121GLM 4.7Zhipu AI75.10-33fragile
122GLM 4.6Zhipu AI75.10-21fragile
123Claude Opus 4.1 (batch)Anthropic75.10+256preliminary
124GLM 4.5Zhipu AI75.10-49fragile
125Qwen3.7 MaxAlibaba74.80+71fragile
126Qwen3.5 Plus 2026-04-20Alibaba74.80+83fragile
127Qwen3.6 Max PreviewAlibaba74.80-51fragile
128Claude Sonnet 4Anthropic74.40-50fragile
129o1OpenAI74.40-50fragile
130o1 (batch)OpenAI74.40+249preliminary
131MiniMax M3MiniMax74.30-51fragile
132o3 Mini High (batch)OpenAI74.30+247preliminary
133o3 Mini (batch)OpenAI74.30+246preliminary
134R1DeepSeek74.20-52fragile
135MiniMax M3 (batch)MiniMax74.10+244preliminary
136Qwen3.6 PlusAlibaba74.10-55fragile
137Inklingthinkingmachines74.00-54fragile
138Hy3Tencent74.00-61fragile
139o1-proOpenAI73.60-55fragile
140o1-pro (batch)OpenAI73.60+239preliminary
141MiMo-V2.5Xiaomi73.00-56fragile
142Gemma 4 26B A4B Google73.00-56fragile
143Gemma 4 26B A4B (free)Google73.00-56fragile
144Inkling (batch)thinkingmachines72.9+1+235preliminary
145GLM 5V TurboZhipu AI72.3+1-56fragile
146o4 Mini HighOpenAI72.1+1-56fragile
147MiniMax M2.1MiniMax72.0+1-44fragile
148MiniMax M2MiniMax72.0+1-57fragile
149GPT-4o (batch)OpenAI72.0+1+230preliminary
150DeepSeek V3.2 ExpDeepSeek71.8+1-48fragile
151DeepSeek V3.1DeepSeek71.8+1-45fragile
152DeepSeek V3 0324DeepSeek71.8+1-60fragile
153Mistral Medium 3.5Mistral AI71.7+1-60fragile
154Inkling Smallthinkingmachines71.6-10-60fragile
155GPT-4o (2024-11-20)OpenAI71.20+4stable
156GPT-4o (2024-08-06)OpenAI71.20-61fragile
157GPT-4oOpenAI71.20-61fragile
158GPT-4o (2024-05-13)OpenAI71.20-61fragile
159MiniMax M1MiniMax70.80-60fragile
160GLM 4.5 AirZhipu AI70.70-60fragile
161Claude Haiku 4.5Anthropic69.50-56fragile
162Claude Haiku 4.5 (batch)Anthropic69.50+217preliminary
163DeepSeek V3DeepSeek69.50-56fragile
164Qwen3 VL 235B A22B ThinkingAlibaba69.40-47fragile
165Qwen3 VL 235B A22B InstructAlibaba69.40-57fragile
166DeepSeek V3.1 TerminusDeepSeek69.30-57fragile
167GPT-4o-miniOpenAI69.30-57fragile
168MiniMax M2-herMiniMax69.10-56fragile
169Qwen3.5-FlashAlibaba68.60-56fragile
170Hy3 previewTencent68.20-59fragile
171Qwen3.5 Plus 2026-02-15Alibaba68.20+55fragile
172Qwen3 Max ThinkingAlibaba68.20-58fragile
173Llama 4 MaverickMeta67.90-58fragile
174GPT-4.1OpenAI67.70-58fragile
175GPT-4.1 (batch)OpenAI67.70+204preliminary
176Qwen3 MaxAlibaba67.40-58fragile
177Mistral Large 3 2512Mistral AI67.00-58fragile
178Qwen3 Next 80B A3B ThinkingAlibaba66.90-41fragile
179Qwen3 Next 80B A3B InstructAlibaba66.90-59fragile
180Llama 3.3 70B InstructMeta66.80-59fragile
181GPT-4 TurboOpenAI66.70-59fragile
182Qwen3.5-9BAlibaba66.50-59fragile
183Step 3.5 FlashStepFun66.10-59fragile
184Mistral Large 2407Mistral AI65.90-33fragile
185Mistral LargeMistral AI65.90-60fragile
186Composer 2Cursor65.70-60fragile
187Composer 2 FastCursor65.70-60fragile
188GLM 4.6VZhipu AI65.50-60fragile
189Qwen3 235B A22B Thinking 2507Alibaba65.50-60fragile
190Llama 3.1 70B InstructMeta65.30-60fragile
191GPT-4 Turbo PreviewOpenAI64.80-45fragile
192GPT-4OpenAI64.80-61fragile
193Qwen3 235B A22B Instruct 2507Alibaba64.70-61fragile
194Qwen3 30B A3B Thinking 2507Alibaba64.10-59fragile
195Qwen3 30B A3B Instruct 2507Alibaba64.10+65fragile
196Qwen3 30B A3BAlibaba64.10-63fragile
197o3 Mini HighOpenAI64.10-63fragile
198Trinity Large Thinkingarcee-ai63.90-62fragile
199GPT-5 MiniOpenAI63.90-61fragile
200GLM 4.7 FlashZhipu AI63.70-61fragile
201Mixtral 8x22B InstructMistral AI63.40-61fragile
202GLM 4.5VZhipu AI62.50-61fragile
203GPT-4o-mini (batch)OpenAI62.50+176preliminary
204GPT-4 Turbo (batch)OpenAI61.50+175preliminary
205Mercury 2Inception61.00-63fragile
206Qwen3 8BAlibaba61.00-63fragile
207Nova 2 LiteAmazon60.50-63fragile
208Phi 4Microsoft60.20-63fragile
209Kimi K2.5Moonshot AI59.10-62fragile
210GPT-4.1 Mini (batch)OpenAI58.90+169preliminary
211GPT-4.1 Nano (batch)OpenAI58.90+168preliminary
212gpt-oss-20bOpenAI57.40-64fragile
213gpt-oss-20b (free)OpenAI57.40-64fragile
214GPT-4o-mini (2024-07-18)OpenAI56.50-64fragile
215Granite 4.1 8BIBM55.30-63fragile
216Olmo 3 32B ThinkAllen AI54.90-63fragile
217Llama 4 ScoutMeta54.70-63fragile
218Kimi K2.7 CodeMoonshot AI54.10-63fragile
219Kimi K2.7 Code (batch)Moonshot AI54.10+160preliminary
220Qwen3 235B A22BAlibaba54.00-64fragile
221GPT-4.1 MiniOpenAI53.60-64fragile
222Kimi K2 ThinkingMoonshot AI53.30-64fragile
223Kimi K2 0905Moonshot AI52.40-63fragile
224Kimi K2 0711Moonshot AI51.40-63fragile
225Claude 3 HaikuAnthropic51.30-63fragile
226Command ACohere50.80-63fragile
227Command R (08-2024)Cohere48.7+1-62fragile
228Command R+ (08-2024)Cohere48.70-63fragile
229GPT-5 NanoOpenAI46.90-63fragile
230Llama 3.1 8B InstructMeta44.50-63fragile
231GPT-4.1 NanoOpenAI42.10-63fragile
232R1 Distill Llama 70BDeepSeek40.70-63fragile
233gpt-oss-120bOpenAI40.40-63fragile
234Ling 3.0 Tiny (free)inclusionai40.00+145preliminary
235DeepSeek V4 Flash Latest~deepseek40.00-64fragile
236DeepSeek V4 Flash 0731DeepSeek40.00-64fragile
237Qwen3.7 FlashAlibaba40.00-64fragile
238Claude Opus 5 (batch)Anthropic40.00+141preliminary
239Ling-3.0-flashinclusionai40.00+140preliminary
240Laguna S 2.1poolside40.00-63fragile
241Laguna S 2.1 (free)poolside40.00-63fragile
242LongCat 2.0Meituan40.00-63fragile
243Kimi K3Moonshot AI40.00-63fragile
244KAT-Coder-Air V2.5Kuaishou40.00-63fragile
245KAT-Coder-Pro V2.5Kuaishou40.00-63fragile
246Grok Latest~x-ai40.00-63fragile
247Aion-3.0-Miniaion-labs40.00-63fragile
248Aion-3.0aion-labs40.00-63fragile
249Laguna XS 2.1poolside40.00-63fragile
250Laguna XS 2.1 (free)poolside40.00-63fragile
251Claude Sonnet 5 (batch)Anthropic40.00+128preliminary
252Fugu Ultrasakana40.00-63fragile
253North Mini Code (free)Cohere40.00-63fragile
254Claude Fable Latest~anthropic40.00-63fragile
255Nemotron 3.5 Content Safety (free)NVIDIA40.00-63fragile
256Nemotron 3 UltraNVIDIA40.00-63fragile
257Nemotron 3 Ultra (batch)NVIDIA40.00+122preliminary
258Nemotron 3 Ultra (free)NVIDIA40.00-64fragile
259Step 3.7 FlashStepFun40.00-64fragile
260Grok Build 0.1xAI40.00+119preliminary
261Perceptron Mk1perceptron40.00-63fragile
262Ring-2.6-1Tinclusionai40.00-63fragile
263GPT Chat LatestOpenAI40.00-63fragile
264Nemotron 3 Nano Omni (free)NVIDIA40.00-63fragile
265Anthropic Claude Haiku Latest~anthropic40.00-63fragile
266OpenAI GPT Mini Latest~openai40.00-63fragile
267Google Gemini Pro Latest~google40.00-63fragile
268MoonshotAI Kimi Latest~moonshotai40.00-63fragile
269Google Gemini Flash Latest~google40.00-63fragile
270Anthropic Claude Sonnet Latest~anthropic40.00-63fragile
271OpenAI GPT Latest~openai40.00-63fragile
272Qwen3.6 FlashAlibaba40.00-62fragile
273Qwen3.6 35B A3BAlibaba40.00-62fragile
274Qwen3.6 27BAlibaba40.00-62fragile
275DeepSeek V4 Flash 0423DeepSeek40.00+104preliminary
276Ling-2.6-1Tinclusionai40.00-63fragile
277Ling-2.6-flashinclusionai40.00-63fragile
278Claude Opus Latest~anthropic40.00-63fragile
279Lyria 3 Pro PreviewGoogle40.00-63fragile
280Lyria 3 Clip PreviewGoogle40.00-63fragile
281KAT-Coder-Pro V2Kuaishou40.00-63fragile
282Reka Edgerekaai40.00-63fragile
283Mistral Small 4Mistral AI40.00-63fragile
284Nemotron 3 SuperNVIDIA40.00-63fragile
285Nemotron 3 Super (free)NVIDIA40.00-63fragile
286Seed-2.0-LiteByteDance40.00-63fragile
287Seed-2.0-MiniByteDance40.00-63fragile
288Aion-2.0aion-labs40.00-63fragile
289Qwen3 Coder NextAlibaba40.00-62fragile
290Solar Pro 3Upstage40.00-62fragile
291Palmyra X5Writer40.00-62fragile
292GPT AudioOpenAI40.00-62fragile
293GPT Audio MiniOpenAI40.00-62fragile
294Falcon-H1-Arabic 34B InstructTII40.00+1stable
295Falcon-H1-Arabic 7B InstructTII40.00+1stable
296Falcon-H1-Arabic 3B InstructTII40.00+1stable
297Seed 1.6 FlashByteDance40.00-65fragile
298Seed 1.6ByteDance40.00-65fragile
299Nemotron 3 Nano 30B A3BNVIDIA40.00-65fragile
300Nemotron 3 Nano 30B A3B (free)NVIDIA40.00-65fragile

服务商分布

OpenAI
86 models
Alibaba
31 models
Anthropic
28 models
Google
27 models
Zhipu AI
13 models
DeepSeek
12 models
NVIDIA
9 models
MiniMax
8 models
Moonshot AI
8 models
Mistral AI
6 models
xAI
5 models
Meta
5 models
inclusionai
5 models
Cohere
4 models
poolside
4 models
~anthropic
4 models
ByteDance
4 models
thinkingmachines
3 models
Kuaishou
3 models
aion-labs
3 models
TII
3 models
meta
2 models
Xiaomi
2 models
Tencent
2 models
StepFun
2 models
Cursor
2 models
~openai
2 models
~google
2 models
arcee-ai
1 model
Inception
1 model
Amazon
1 model
Microsoft
1 model
IBM
1 model
Allen AI
1 model
~deepseek
1 model
Meituan
1 model
~x-ai
1 model
sakana
1 model
perceptron
1 model
~moonshotai
1 model
rekaai
1 model
Upstage
1 model
Writer
1 model

相关

Frequently Asked Questions

Models are ranked using a benchmark-driven score from 0 to 100 where benchmark performance (90%) from LMArena, MMLU, GPQA, HumanEval, SWE-bench, and 15+ standardized evaluations is the primary signal. Capabilities and context window serve as tiebreakers (10%). Scores update hourly from live API data across 290+ coding models.

The 24h column shows how many positions a model moved up or down in the last 24 hours, while the 7d column shows the change over the past week. Green values with a + indicate rank improvements, red values indicate drops, and 0 means the rank was unchanged.

The state classification reflects a model's ranking consistency. "Stable" means the model has maintained its position reliably. "Held" indicates it is holding steady but with some variance. "Fragile" means the model's rank is fluctuating and may shift significantly. "Preliminary" is assigned to newly tracked models without enough history.

Coding AI Tracker — Daily Performance & Rankings | LM Market Cap