Math benchmark

LMArena Math leaderboard

As of October 2026, Gemini 4 Argon has the highest published LMArena Math score on Noometry at 1536, out of 285 models with results.

Last verified

About LMArena Math

LMArena ratings restricted to math prompts.

Category
Math
Introduced
2024
Format
Pairwise human votes
Unit
Arena rating (Bradley–Terry)
Official site
lmarena.ai

Top 15 models

Top models on LMArena Math
  1. Gemini 4 Argon 1536
  2. Claude Opus 5 1531
  3. Gemini 3.8 Flash 1528
  4. Claude Fable 5.1 1525
  5. Claude Fable 5 1519
  6. Claude Opus 4.6 1519
  7. Claude Sonnet 5.5 1510
  8. Gemini 3.7 Flash 1507
  9. Claude Opus 5.5 1506
  10. Gemini 3.6 Flash 1505
  11. Gemini 3.5 Flash 1504
  12. GLM-5.3-Flash 1500
  13. Qwen3.8 Max 1499
  14. Claude Opus 4.7 1499
  15. MiMo-V2.6-Pro 1494

Sponsored placements are available on pages like this one. Advertise on Noometry

All results

LMArena Math results by model
#ModelProviderRatingSettingSourceDate
1Gemini 4 Argon Google1536highLMArena2026-10-08
2Claude Opus 5 Anthropic1531highLMArena2026-10-08
3Gemini 3.8 Flash Google1528highLMArena2026-10-08
4Claude Fable 5.1 Anthropic1525LMArena2026-10-08
5Claude Fable 5 Anthropic1519highLMArena2026-10-08
6Claude Opus 4.6 Anthropic1519highLMArena2026-10-08
7Claude Sonnet 5.5 Anthropic1510xhighLMArena2026-10-08
8Gemini 3.7 Flash Google1507highLMArena2026-10-08
9Claude Opus 5.5 Anthropic1506highLMArena2026-10-08
10Gemini 3.6 Flash Google1505highLMArena2026-10-08
11Gemini 3.5 Flash Google1504highLMArena2026-10-08
12GLM-5.3-Flash Z.ai (Zhipu)1500LMArena2026-10-08
13Qwen3.8 Max Alibaba (Qwen)1499LMArena2026-10-08
14Claude Opus 4.7 Anthropic1499highLMArena2026-10-08
15MiMo-V2.6-Pro Xiaomi1494LMArena2026-10-08
16Muse Spark 1.3 Meta1494LMArena2026-10-08
17Kimi K3 Moonshot AI1491LMArena2026-10-08
18Qwen3.7 Max Alibaba (Qwen)1490LMArena2026-10-08
19GLM-5.3 Z.ai (Zhipu)1489LMArena2026-10-08
20Mistral Large 4 Mistral AI1488LMArena2026-10-08
21GPT-5.4 OpenAI1488highLMArena2026-10-08
22Claude Opus 4.8 Anthropic1487highLMArena2026-10-08
23GPT-5.5 OpenAI1486LMArena2026-10-08
24Gemini 3.1 Pro Preview Google1485LMArena2026-10-08
25Muse Spark 1.1 Meta1483LMArena2026-10-08
26GLM-5.2 Z.ai (Zhipu)1482LMArena2026-10-08
27ERNIE 5.1 Baidu1481LMArena2026-10-08
28MiMo-V2.5-Pro Xiaomi1481LMArena2026-10-08
29Inkling Thinking Machines Lab1479LMArena2026-10-08
30DeepSeek V4.1 Flash DeepSeek1477LMArena2026-10-08
31Gemini 3 Pro Google1476LMArena2026-10-08
32Kimi K2.6 Moonshot AI1475LMArena2026-10-08
33Hy3 Tencent1475LMArena2026-10-08
34Qwen3.5 Max Preview Alibaba (Qwen)1474LMArena2026-10-08
35GPT-5.6 Sol OpenAI1474xhighLMArena2026-10-08
36GLM-5.1 Z.ai (Zhipu)1473LMArena2026-10-08
37Gemini 3 Flash Preview Google1473LMArena2026-10-08
38Muse Spark 1.2 Meta1471muse-spark-1.2 (xHigh)LMArena2026-10-08
39Gemma 4 26B A4B IT Google1470LMArena2026-10-08
40Kimi K2.5 Moonshot AI1470thinkingLMArena2026-10-08
41Step 5 Preview StepFun1470Step 5 PreviewLMArena2026-10-08
42MiMo-V2.6-Flash Xiaomi1468LMArena2026-10-08
43Claude Sonnet 5 Anthropic1467highLMArena2026-10-08
44Qwen3.7 Plus Alibaba (Qwen)1466LMArena2026-10-08
45GPT-5.6 Terra OpenAI1466xhighLMArena2026-10-08
46GPT-6 Astra OpenAI1465LMArena2026-10-08
47Qwen3.6 Max Preview Alibaba (Qwen)1465LMArena2026-10-08
48Gemma 4 31B IT Google1465LMArena2026-10-08
49GPT-6.1 Sol OpenAI1464LMArena2026-10-08
50Claude Opus 4.5 Anthropic1463LMArena2026-10-08
51Claude Sonnet 4.6 Anthropic1462LMArena2026-10-08
52Grok 4.5 xAI1459LMArena2026-10-08
53Inkling-Small Thinking Machines Lab1459Inkling SmallLMArena2026-10-08
54GPT-5.6 Luna OpenAI1458xhighLMArena2026-10-08
55Nemotron 3 Ultra NVIDIA1457LMArena2026-10-08
56Qwen3.8 27B Alibaba (Qwen)1456LMArena2026-10-08
57Grok 4.20 (Non-Reasoning) xAI1455LMArena2026-10-08
58Muse Spark Meta1455LMArena2026-10-08
59DeepSeek V4 Pro DeepSeek1455LMArena2026-10-08
60Qwen3.5 397B-A17B Alibaba (Qwen)1454LMArena2026-10-08
61Qwen3.6 Plus Alibaba (Qwen)1450LMArena2026-10-08
62Gemini 2.5 Pro Google1450LMArena2026-10-08
63Claude Sonnet 4.5 Anthropic1449LMArena2026-10-08
64MiMo-V2-Pro Xiaomi1447LMArena2026-10-08
65GPT-5.1 OpenAI1447highLMArena2026-10-08
66Qwen3 Max Alibaba (Qwen)1446LMArena2026-10-08
67Grok 4.20 Multi-Agent xAI1442LMArena2026-10-08
68Kimi K2.5 Instant Moonshot AI1442LMArena2026-10-08
69Longcat Flash Chat Meituan1442LMArena2026-10-08
70GLM-5V-Turbo Z.ai (Zhipu)1441LMArena2026-10-08
71Qwen3-Next 80B-A3B Instruct Alibaba (Qwen)1440LMArena2026-10-08
72GPT-5.2 OpenAI1440highLMArena2026-10-08
73GLM-5 Z.ai (Zhipu)1440LMArena2026-10-08
74Seed 2.0 Pro ByteDance Seed1439LMArena2026-10-08
75Amazon Nova Experimental Chat 26 02 10 Amazon1438LMArena2026-10-08
76ERNIE 5.0 0110 Baidu1437LMArena2026-10-08
77MiMo-V2.5 Xiaomi1436LMArena2026-10-08
78DeepSeek-V3.2-Exp DeepSeek1435LMArena2026-10-08
79Gemini 3.5 Flash Lite Google1433LMArena2026-10-08
80GLM-4.6 Z.ai (Zhipu)1432LMArena2026-10-08
81Qwen3 235B-A22B Alibaba (Qwen)1432LMArena2026-10-08
82Qwen3.5 122B-A10B Alibaba (Qwen)1432LMArena2026-10-08
83Mistral Medium 3.5 Mistral AI1431LMArena2026-10-08
84Claude Opus 4.1 Anthropic1431thinking-16kLMArena2026-10-08
85Amazon Nova Experimental Chat 11 10 Amazon1431LMArena2026-10-08
86MiMo-V2-Omni Xiaomi1430LMArena2026-10-08
87MiniMax-M3 MiniMax1429LMArena2026-10-08
88Qwen3.5 27B Alibaba (Qwen)1429LMArena2026-10-08
89Kimi K2 Thinking Turbo Moonshot AI1429LMArena2026-10-08
90Gemini 3.1 Flash Lite Google1428LMArena2026-10-08
91GLM-4.5 Z.ai (Zhipu)1427LMArena2026-10-08
92DeepSeek V4 Flash DeepSeek1427LMArena2026-10-08
93o3 OpenAI1426LMArena2026-10-08
94Qwen3-VL 235B-A22B Alibaba (Qwen)1426LMArena2026-10-08
95Amazon Nova Experimental Chat 10 20 Amazon1425LMArena2026-10-08
96Grok 4.6 xAI1423highLMArena2026-10-08
97GLM-4.7 Z.ai (Zhipu)1423LMArena2026-10-08
98Grok 4 xAI1422LMArena2026-10-08
99Grok 4.1 xAI1422thinkingLMArena2026-10-08
100MiniMax-M2.7 MiniMax1420LMArena2026-10-08
101DeepSeek-V3.1 DeepSeek1420LMArena2026-10-08
102GPT-5.5 Instant OpenAI1420LMArena2026-10-08
103Grok 4 Fast xAI1419LMArena2026-10-08
104Amazon Nova Experimental Chat 12 10 Amazon1419LMArena2026-10-08
105GPT-5.4 mini OpenAI1419highLMArena2026-10-08
106Muse Glimmer Meta1417LMArena2026-10-08
107Solar Pro4 Upstage1416LMArena2026-10-08
108GPT-6 Luna OpenAI1416LMArena2026-10-08
109Gemini 2.5 Flash Google1415LMArena2026-10-08
110Mistral Large 3 Mistral AI1414LMArena2026-10-08
111Hunyuan T1 20250711 Tencent1414LMArena2026-10-08
112GPT-4.5 OpenAI1412LMArena2026-10-08
113Step 3.5 Flash StepFun1408LMArena2026-10-08
114Mistral Medium Mistral AI1408LMArena2026-10-08
115Grok 4.1 Fast xAI1408LMArena2026-10-08
116GPT-5 OpenAI1407LMArena2026-10-08
117Grok 4.7 xAI1407xhighLMArena2026-10-08
118Qwen3.5-Flash Alibaba (Qwen)1407LMArena2026-10-08
119Chatgpt 4o Latest 20250326 OpenAI1407LMArena2026-10-08
120GPT-5.4 nano OpenAI1406highLMArena2026-10-08
121Qwen3.5 35B-A3B Alibaba (Qwen)1404LMArena2026-10-08
122Amazon Nova Experimental Chat 26 01 10 Amazon1404LMArena2026-10-08
123GPT-6 Sol OpenAI1402LMArena2026-10-08
124DeepSeek-V3.1-Terminus DeepSeek1402LMArena2026-10-08
125DeepSeek-R1 DeepSeek1400LMArena2026-10-08
126Qwen3 32B Alibaba (Qwen)1399LMArena2026-10-08
127Kimi K2 (Jul 2025) Moonshot AI1397LMArena2026-10-08
128MiniMax-M2.1 MiniMax1397LMArena2026-10-08
129GLM-4.5-Air Z.ai (Zhipu)1396LMArena2026-10-08
130Claude Haiku 4.5 Anthropic1396LMArena2026-10-08
131MiMo-V2-Flash Xiaomi1396LMArena2026-10-08
132o3-mini OpenAI1396highLMArena2026-10-08
133Qwen3-30B-A3B Alibaba (Qwen)1394LMArena2026-10-08
134Nvidia Llama 3.3 Nemotron Super 49b v1.5 NVIDIA1392LMArena2026-10-08
135Grok 3 xAI1391LMArena2026-10-08
136Claude Opus 4 Anthropic1390thinking-16kLMArena2026-10-08
137GPT-5.3 Chat OpenAI1389LMArena2026-10-08
138gpt-oss-120b OpenAI1389LMArena2026-10-08
139o4-mini OpenAI1389LMArena2026-10-08
140o1 OpenAI1388LMArena2026-10-08
141Grok 4.3 xAI1388LMArena2026-10-08
142Grok-3 mini xAI1386highLMArena2026-10-08
143Nemotron 3 Super NVIDIA1380LMArena2026-10-08
144GPT-5 Mini OpenAI1378highLMArena2026-10-08
145MiniMax-M2.5 MiniMax1378LMArena2026-10-08
146Claude Sonnet 4 Anthropic1375thinking-32kLMArena2026-10-08
147Gemini 2.5 Flash-Lite Google1373no-thinkingLMArena2026-10-08
148DeepSeek-V3 DeepSeek1373LMArena2026-10-08
149Nemotron 3 Nano 30B A3B NVIDIA1371LMArena2026-10-08
150GPT-4.1 OpenAI1370LMArena2026-10-08
151Qwen2.5-Max Alibaba (Qwen)1369LMArena2026-10-08
152Trinity Large Thinking Arcee AI1366thinkingLMArena2026-10-08
153Step 3 StepFun1366LMArena2026-10-08
154Qwen3-Coder 480B-A35B Instruct Alibaba (Qwen)1365LMArena2026-10-08
155MiniMax M1 MiniMax1361LMArena2026-10-08
156Llama 3.1 Nemotron Ultra 253b v1 NVIDIA1360LMArena2026-10-08
157QwQ-32B Alibaba (Qwen)1359LMArena2026-10-08
158Nova 2 Lite Amazon1359LMArena2026-10-08
159Nemotron 3.5 Lightning NVIDIA1359LMArena2026-10-08
160Hunyuan Turbos 20250226 Tencent1359LMArena2026-10-08
161o1-mini OpenAI1358LMArena2026-10-08
162GLM-4.7-Flash Z.ai (Zhipu)1355LMArena2026-10-08
163GLM-4.5V Z.ai (Zhipu)1354LMArena2026-10-08
164MiniMax-M2 MiniMax1352LMArena2026-10-08
165Gemini 2.0 Flash (Feb 2025) Google1352LMArena2026-10-08
166GPT-4.1 mini OpenAI1343LMArena2026-10-08
167Mistral Small Mistral AI1341LMArena2026-10-08
168Claude 3.7 Sonnet Anthropic1337thinking-32kLMArena2026-10-08
169Qwen Plus Alibaba (Qwen)1326LMArena2026-10-08
170Step 1o Turbo 202506 StepFun1318LMArena2026-10-08
171GPT-5 Nano OpenAI1317highLMArena2026-10-08
172gpt-oss-20b OpenAI1317LMArena2026-10-08
173Olmo 3 32b Think Allen Institute for AI (Ai2)1316LMArena2026-10-08
174Gemini 1.5 Pro (May 2024) Google1315LMArena2026-10-08
175Gemma 3 27B Google1312LMArena2026-10-08
176Granite 4.1 8b IBM1312LMArena2026-10-08
177Gemini 2.0 Flash-Lite Google1309LMArena2026-10-08
178Gemma 3 12B Google1307LMArena2026-10-08
179Claude 3.5 Sonnet Anthropic1307LMArena2026-10-08
180Olmo 3.1 32b Instruct Allen Institute for AI (Ai2)1305LMArena2026-10-08
181Olmo 3.1 32b Think Allen Institute for AI (Ai2)1305LMArena2026-10-08
182Step 2 16k Exp 202412 StepFun1304LMArena2026-10-08
183Command A Cohere1300LMArena2026-10-08
184Yi-Lightning 01.AI1300LMArena2026-10-08
185Llama 4 Maverick Meta1299LMArena2026-10-08
186Qwen2.5 Plus 1127 Alibaba (Qwen)1298LMArena2026-10-08
187DeepSeek-V2.5 (Sep 2024) DeepSeek1288LMArena2026-10-08
188Llama 4 Scout Meta1287LMArena2026-10-08
189GPT-4o OpenAI1285LMArena2026-10-08
190Grok-2 (Dec 2024) xAI1283LMArena2026-10-08
191Qwen2.5 72B Instruct Alibaba (Qwen)1283LMArena2026-10-08
192Llama 3.1-405B Meta1281LMArena2026-10-08
193Hunyuan Large 2025 02 10 Tencent1281LMArena2026-10-08
194Qwen Max Alibaba (Qwen)1275LMArena2026-10-08
195GPT-4.1 nano OpenAI1274LMArena2026-10-08
196Hunyuan Standard 2025 02 10 Tencent1274LMArena2026-10-08
197Hunyuan Turbo 0110 Tencent1273LMArena2026-10-08
198Claude 3 Opus Anthropic1273LMArena2026-10-08
199Gemini Advanced 0514 Google1272LMArena2026-10-08
200GPT-4 Turbo OpenAI1272LMArena2026-10-08
201Llama 3.1 Nemotron 70b Instruct NVIDIA1271LMArena2026-10-08
202GPT-4 OpenAI1269LMArena2026-10-08
203Gemini 1.5 Flash (May 2024) Google1269LMArena2026-10-08
204Hunyuan Large Vision Tencent1268LMArena2026-10-08
205GPT-4o mini OpenAI1267LMArena2026-10-08
206Llama-3.3-70B-Instruct Meta1267LMArena2026-10-08
207Grok 2 Mini 2024 08 13 xAI1265LMArena2026-10-08
208Mistral Large Mistral AI1262LMArena2026-10-08
209Mistral Small 3.1 Mistral AI1262LMArena2026-10-08
210Llama 3.1-70B Meta1252LMArena2026-10-08
211Amazon Nova Pro Amazon1252LMArena2026-10-08
212Gemma 3n E4b IT Google1251LMArena2026-10-08
213Qwen2.5-Coder-32B Alibaba (Qwen)1251LMArena2026-10-08
214Magistral Medium Mistral AI1250LMArena2026-10-08
215Granite 4.0 H Small IBM1247LMArena2026-10-08
216Phi-4 Microsoft1246LMArena2026-10-08
217Claude 3.5 Haiku Anthropic1244LMArena2026-10-08
218Tulu 3 (Tülu 3) 70B Allen Institute for AI (Ai2)1242LMArena2026-10-08
219Deepseek Coder v2 DeepSeek1241LMArena2026-10-08
220Mistral Small 3 Mistral AI1240LMArena2026-10-08
221Gemma 3 4B Google1239LMArena2026-10-08
222Qwen2-72B Alibaba (Qwen)1235LMArena2026-10-08
223Llama 3.1 Nemotron 51b Instruct NVIDIA1230LMArena2026-10-08
224Amazon Nova Lite Amazon1227LMArena2026-10-08
225Llama 3-70B Meta1218LMArena2026-10-08
226Nemotron 4 340b Instruct NVIDIA1216LMArena2026-10-08
227Claude 3 Sonnet Anthropic1213LMArena2026-10-08
228Gemma 2 27B Google1212LMArena2026-10-08
229Olmo 2 0325 32b Instruct Allen Institute for AI (Ai2)1208LMArena2026-10-08
230Gemini 1.5 Flash 8B Google1207LMArena2026-10-08
231Amazon Nova Micro Amazon1206LMArena2026-10-08
232C4ai Aya Expanse 32b Cohere1200LMArena2026-10-08
233Llama 3.1 Tulu 3 8b Allen Institute for AI (Ai2)1195LMArena2026-10-08
234Ministral 8B Mistral AI1188LMArena2026-10-08
235Claude 3 Haiku Anthropic1188LMArena2026-10-08
236Command R+ Cohere1188LMArena2026-10-08
237Qwen1.5-110B Alibaba (Qwen)1185LMArena2026-10-08
238Mixtral 8x22B Mistral AI1184LMArena2026-10-08
239Gemma 2 9B Google1183LMArena2026-10-08
240Yi-1.5-34B 01.AI1182LMArena2026-10-08
241Mistral Mistral AI1180mediumLMArena2026-10-08
242Llama 3.1-8B Meta1179LMArena2026-10-08
243Phi 3 Medium 4k Instruct Microsoft1173LMArena2026-10-08
244C4ai Aya Expanse 8b Cohere1168LMArena2026-10-08
245Qwen1.5-72B Alibaba (Qwen)1164LMArena2026-10-08
246Granite 3.1 2b Instruct IBM1159LMArena2026-10-08
247Qwen1.5-32B Alibaba (Qwen)1155LMArena2026-10-08
248Command R Cohere1155LMArena2026-10-08
249Phi 3 Mini 4k Instruct June 2024 Microsoft1152LMArena2026-10-08
250Granite 3.1 8b Instruct IBM1152LMArena2026-10-08
251Llama 3-8B Meta1151LMArena2026-10-08
252Phi 3 Small 8k Instruct Microsoft1151LMArena2026-10-08
253Mixtral 8x7B Mistral AI1147LMArena2026-10-08
254DBRX Databricks1145LMArena2026-10-08
255Granite 3.0 8b Instruct IBM1143LMArena2026-10-08
256GPT-3.5-turbo OpenAI1142LMArena2026-10-08
257Gemma 2 2b IT Google1135LMArena2026-10-08
258Gemini 1.0 Pro Google1132LMArena2026-10-08
259Llama 3.2 3B Meta1126LMArena2026-10-08
260Qwen1.5-14B Alibaba (Qwen)1125LMArena2026-10-08
261Granite 3.0 2b Instruct IBM1117LMArena2026-10-08
262Wizardlm 70b Microsoft1116LMArena2026-10-08
263Yi-34B 01.AI1114LMArena2026-10-08
264Phi 3 Mini 4k Instruct Microsoft1111LMArena2026-10-08
265DeepSeek LLM 67B DeepSeek1108LMArena2026-10-08
266Gemma 1.1 7b IT Google1107LMArena2026-10-08
267Llama 2-70B Meta1091LMArena2026-10-08
268Phi 3 Mini 128k Instruct Microsoft1089LMArena2026-10-08
269Llama 3.2 1B Meta1086LMArena2026-10-08
270Mistral 7B Mistral AI1085LMArena2026-10-08
271Qwen1.5-7B Alibaba (Qwen)1080LMArena2026-10-08
272Llama2 70b Steerlm Chat NVIDIA1072LMArena2026-10-08
273Qwen-14B Alibaba (Qwen)1068LMArena2026-10-08
274Gemma 7B Google1066LMArena2026-10-08
275Llama 2-13B Meta1065LMArena2026-10-08
276Codellama 34b Instruct Meta1056LMArena2026-10-08
277Palm 2 Google1049LMArena2026-10-08
278Gemma 1.1 2b IT Google1047LMArena2026-10-08
279Llama 2-7B Meta1042LMArena2026-10-08
280Qwen1.5 4b Chat Alibaba (Qwen)1026LMArena2026-10-08
281Olmo 7b Instruct Allen Institute for AI (Ai2)1018LMArena2026-10-08
282Wizardlm 13b Microsoft1017LMArena2026-10-08
283Gemma 2B Google1009LMArena2026-10-08
284Dolly 2.0-12b Databricks871LMArena2026-10-08
285Llama 13b Meta838LMArena2026-10-08

Compare the leaders

Other math benchmarks

Frequently asked questions

What does LMArena Math measure?

LMArena ratings restricted to math prompts.

Which model has the highest LMArena Math score?

As of October 2026, Gemini 4 Argon has the highest published LMArena Math score on Noometry at 1536, out of 285 models with results.

What is the best open-weight model on LMArena Math?

GLM-5.3-Flash has the highest LMArena Math rating among open-weight models at 1500, ranking 12 of 285 overall.