1 A Gemini 4 Argon (High)provisional Google 1,506 ±19 100.0 — A ›
2 A Claude Fable 5.1provisional Anthropic 1,496 ±13 99.6 +47 A ›
3 A Gemini 3.7 Flashprovisional Google 1,490 ±10 99 +81 A ›
4 A Gemini 3.8 Flashprovisional Google 1,486 ±9 99 +84◆ A ›
5 A Claude Opus 5provisional Anthropic 1,484 ±9 98 +59 A ›
6 A Claude Fable 5provisional Anthropic 1,483 ±8 98 +33 A ›
7 A Claude Opus 5.5provisional Anthropic 1,478 ±19 97 +76◆ A ›
8 A Claude Opus 4.6provisional Anthropic 1,476 ±6 97 +41 A ›
9 A Gemini 3.1 Pro Previewprovisional Google 1,476 ±5 97 +42 A ›
10 A Claude Opus 4.7provisional Anthropic 1,473 ±7 96 +35 A ›
11 A gemini-3-proprovisional Google 1,472 ±7 96 +33 A ›
12 A Gemini 3.5 Flashprovisional Google 1,472 ±7 95 +49 A ›
13 A Muse Spark 1.3provisional Meta 1,470 ±13 95 +74 A ›
14 A Gemini 3.6 Flashprovisional Google 1,468 ±8 94 +68 A ›
15 A Muse Spark 1.1provisional Meta 1,464 ±8 94 +58 A ›
16 A Kimi K3provisional Moonshot AI · Open weights 1,463 ±9 94 +44 A ›
17 A GPT-5.4provisional OpenAI 1,462 ±7 93 +39 A ›
18 A Gemini 3 Flash Previewprovisional Google 1,462 ±8 93 +57 A ›
19 A Qwen3.8 Max (0902)provisional Alibaba 1,460 ±9 92 +54 A ›
20 A Claude Opus 4.8provisional Anthropic 1,456 ±6 92 +27 A ›
21 A Muse Spark 1.2provisional Meta 1,456 ±21 91 +62 A ›
22 A GPT-5.5provisional OpenAI 1,456 ±6 91 +29 A ›
23 A MiMo-V2.6-Proprovisional Xiaomi · Open weights 1,454 ±19 91 +93◆ A ›
24 A GLM 5.3 Flashprovisional Zhipu AI · Open weights 1,453 ±9 90 +105◆ A ›
25 A qwen3.5-max-previewprovisional Alibaba 1,453 ±10 90 — A ›
26 A Qwen3.7 Maxprovisional Alibaba 1,451 ±25 89 +39 A ›
27 A GPT-5.6 Solprovisional OpenAI 1,451 ±8 89 +40 A ›
28 A GLM 5.2provisional Zhipu AI · Open weights 1,450 ±7 88 +45 A ›
29 A DeepSeek V4.1 Flashprovisional DeepSeek · Open weights 1,450 ±14 88 +92◆ A ›
30 A muse-sparkprovisional Meta 1,449 ±13 88 — A ›
31 A Gemini 2.5 Proprovisional Google 1,446 ±5 87 +5 A ›
32 A ernie-5.1provisional Baidu 1,444 ±8 87 +52 A ›
33 A Grok 4.20 (Reasoning)provisional xAI 1,443 ±6 86 +26 A ›
34 A Grok 4.20 Multi-Agentprovisional xAI 1,442 ±6 86 +24 A ›
35 A Gemma 4 31Bprovisional Google · Open weights 1,442 ±16 85 +58 A ›
36 A GLM 5.3provisional Zhipu AI · Open weights 1,442 ±11 85 +60 A ›
37 A Qwen3.7 Plusprovisional Alibaba 1,442 ±8 84 +52 A ›
38 A GLM 5.1provisional Zhipu AI · Open weights 1,439 ±6 84 +36 A ›
39 A DeepSeek V4 Pro 0423provisional DeepSeek · Open weights 1,439 ±7 84 +33 A ›
40 A MiMo-V2.5-Proprovisional Xiaomi · Open weights 1,437 ±6 83 +80 A ›
41 A Claude Sonnet 5provisional Anthropic 1,437 ±7 83 +32 A ›
42 A dola-seed-2.0-proprovisional ByteDance 1,436 ±6 82 +31 A ›
43 A GPT-5.6 Terraprovisional OpenAI 1,436 ±8 82 +38 A ›
44 A Claude Opus 4.5provisional Anthropic 1,435 ±6 82 +2 A ›
45 A Grok 4.20provisional xAI 1,434 ±9 81 +33 A ›
46 A Grok 4.5provisional xAI 1,434 ±8 81 +35 A ›
47 A ernie-5.0-0110provisional Baidu 1,433 ±7 80 — A ›
48 B Kimi K2.6provisional Moonshot AI · Open weights 1,433 ±8 80 +44 A ›
49 B Gemini 3.5 Flash Liteprovisional Google 1,431 ±8 79 +29 A ›
50 B Claude Sonnet 4.6provisional Anthropic 1,428 ±6 79 +21 A ›
51 B GPT-5.2provisional OpenAI 1,428 ±8 78 +12 A ›
52 B Claude Sonnet 4.5provisional Anthropic 1,425 ±5 78 -2 A ›
53 B Step 5 Previewprovisional StepFun 1,425 ±22 78 +32 A ›
54 B Qwen3.6 Max Previewprovisional Alibaba 1,425 ±20 77 +20 A ›
55 B GPT-6 Astraprovisional OpenAI 1,425 ±14 77 +12 A ›
56 B Kimi K2.5provisional Moonshot AI · Open weights 1,424 ±6 76 +36 A ›
57 B GLM 5provisional Zhipu AI · Open weights 1,424 ±8 76 +23 A ›
58 B GPT-5.1provisional OpenAI 1,424 ±7 76 +18 A ›
59 B GPT-6.1 Solprovisional OpenAI 1,424 ±22 75 +53 A ›
60 B grok-4.1-thinkingprovisional xAI 1,423 ±6 75 -12 A ›
61 B Qwen3.6 Plusprovisional Alibaba 1,423 ±7 74 +34 A ›
62 B Claude Sonnet 5.5provisional Anthropic 1,422 ±22 74 +50 A ›
63 B Gemma 4 26B A4B provisional Google · Open weights 1,420 ±15 73 +54 A ›
64 B Hy3provisional Tencent · Open weights 1,419 ±14 73 +68 A ›
65 B MiniMax M3provisional MiniMax · Open weights 1,418 ±7 72 +50 A ›
66 B Qwen3.5 397B A17Bprovisional Alibaba · Open weights 1,418 ±6 72 +27 A ›
67 B GLM 5V Turboprovisional Zhipu AI 1,418 ±15 72 +14 A ›
68 B GPT-5.6 Lunaprovisional OpenAI 1,418 ±8 71 +80 A ›
69 B DeepSeek V4 Flash 0423provisional DeepSeek · Open weights 1,417 ±7 71 +71 A ›
70 B Inklingprovisional Thinking Machines Lab · Open weights 1,417 ±8 70 +26 A ›
71 B MiMo-V2-Proprovisional Xiaomi 1,417 ±9 70 +31 A ›
72 B Qwen3 Maxprovisional Alibaba 1,416 ±11 70 +2 A ›
73 B Grok 4.6provisional xAI 1,415 ±9 69 +32 A ›
74 B DeepSeek V3.2provisional DeepSeek 1,414 ±7 69 +54 A ›
75 B Claude Opus 4.1provisional Anthropic 1,412 ±6 68 -61 A ›
76 B GLM 4.7provisional Zhipu AI · Open weights 1,412 ±12 68 +20 A ›
77 B R1 0528provisional DeepSeek 1,412 ±12 67 +2 A ›
78 B Gemini 3.1 Flash Lite Previewprovisional Google 1,411 ±6 67 +16 A ›
79 B GLM 4.6provisional Zhipu AI · Open weights 1,409 ±8 66 +26 A ›
80 B GPT-5.4 Miniprovisional OpenAI 1,408 ±7 66 +14 A ›
81 B amazon-nova-experimental-chat-26-02-10provisional Amazon 1,408 ±20 66 — A ›
82 B Qwen3.8 27Bprovisional Alibaba · Open weights 1,407 ±10 65 +46 A ›
83 B Gemini 2.5 Flashprovisional Google 1,407 ±5 65 +7 A ›
84 B grok-3-preview-02-24provisional xAI 1,407 ±9 64 -24 A ›
85 B MiMo-V2.6-Flashprovisional Xiaomi · Open weights 1,407 ±16 64 +98◆ A ›
86 B mistral-large-3provisional Mistral AI 1,405 ±5 64 +25 A ›
87 B GLM 4.5provisional Zhipu AI · Open weights 1,405 ±12 63 +18 A ›
88 B Nemotron 3 Ultraprovisional NVIDIA · Open weights 1,404 ±14 63 +33 A ›
89 B Mistral Medium 3.1provisional Mistral AI 1,404 ±5 62 +19 A ›
90 B MiMo-V2-Omniprovisional Xiaomi 1,403 ±11 62 +46 A ›
91 B grok-4-0709provisional xAI 1,401 ±9 61 -27 A ›
92 B o3provisional OpenAI 1,400 ±7 61 -17 A ›
93 B GPT-6 Solprovisional OpenAI 1,399 ±14 60 +22 A ›
94 B GPT-5provisional OpenAI 1,399 ±10 60 -4 A ›
95 B DeepSeek V3.2 Expprovisional DeepSeek · Open weights 1,398 ±18 60 +46 A ›
96 B Qwen3 VL 235B A22B Instructprovisional Alibaba · Open weights 1,398 ±17 59 +27 A ›
97 B Qwen3 Next 80B A3B Instructprovisional Alibaba · Open weights 1,396 ±12 59 +32 A ›
98 B DeepSeek V3.1provisional DeepSeek · Open weights 1,395 ±15 58 +19 A ›
99 B GPT-5.3 Chat (latest)provisional OpenAI 1,394 ±8 58 -45 A ›
100 B Grok 4.7provisional xAI 1,393 ±16 58 +11 A ›
101 B Qwen3.5-122B-A10Bprovisional Alibaba · Open weights 1,393 ±9 57 +14 A ›
102 B Grok 4.3provisional xAI 1,393 ±6 57 +4 A ›
103 B MiMo-V2.5provisional Xiaomi · Open weights 1,390 ±7 56 +75 A ›
104 B Qwen3 235B A22B Thinking 2507provisional Alibaba 1,389 ±20 56 +12 A ›
105 B Mistral Medium 3.5provisional Mistral AI · Open weights 1,388 ±14 55 -19 A ›
106 B longcat-flash-chat-2602-expprovisional Meituan 1,388 ±16 55 — A ›
107 B gemini-2.5-flash-preview-09-2025provisional Google 1,388 ±9 54 +8 A ›
108 B hunyuan-hy3-previewprovisional Tencent 1,387 ±19 54 — A ›
109 B claude-opus-4-20250514-thinking-16kprovisional Anthropic 1,387 ±10 54 -86 A ›
110 B kimi-k2-thinking-turboprovisional Moonshot AI 1,386 ±6 53 -5 A ›
111 B amazon-nova-experimental-chat-12-10provisional Amazon 1,386 ±20 53 — A ›
112 B Inkling Smallprovisional Thinking Machines Lab · Open weights 1,386 ±9 52 +38 A ›
113 B Qwen3.5-27Bprovisional Alibaba · Open weights 1,385 ±9 52 +14 A ›
114 B MiMo-V2-Flashprovisional Xiaomi · Open weights 1,383 ±6 52 +55◆ A ›
115 B MiniMax M2.1provisional MiniMax · Open weights 1,382 ±10 51 +18 A ›
116 B GPT-6 Lunaprovisional OpenAI 1,382 ±14 51 +80 A ›
117 B Step 3.5 Flashprovisional StepFun · Open weights 1,382 ±6 50 +50 A ›
118 C MiniMax M2.7provisional MiniMax · Open weights 1,380 ±6 50 +32 A ›
119 C muse-glimmerprovisional Meta 1,380 ±23 49 — A ›
120 C Kimi K2 0711provisional Moonshot AI · Open weights 1,378 ±16 49 -22 A ›
121 C Claude Haiku 4.5provisional Anthropic 1,378 ±4 48 -12 A ›
122 C Qwen3.5-Flashprovisional Alibaba 1,376 ±6 48 +38 A ›
123 C GPT-4.1provisional OpenAI 1,374 ±8 48 -56 A ›
124 C Qwen3.5-35B-A3Bprovisional Alibaba · Open weights 1,374 ±8 47 +11 A ›
125 C hunyuan-t1-20250711provisional Tencent 1,372 ±25 47 — A ›
126 C gemini-2.5-flash-lite-preview-09-2025-no-thinkingprovisional Google 1,371 ±7 46 +19 A ›
127 C GLM 4.5 Airprovisional Zhipu AI · Open weights 1,371 ±11 46 -2 A ›
128 C Qwen3 30B A3B Instruct 2507provisional Alibaba · Open weights 1,367 ±12 46 +25 A ›
129 C Qwen3 Coder 480B A35Bprovisional Alibaba · Open weights 1,365 ±12 45 -22 A ›
130 C hunyuan-turbos-20250416provisional Tencent 1,365 ±14 45 — A ›
131 C Qwen3 VL 235B A22B Thinkingprovisional Alibaba · Open weights 1,364 ±19 44 -17 A ›
132 C GPT-5.4 Nanoprovisional OpenAI 1,362 ±7 44 +14 A ›
133 C Gemini 2.5 Flash Liteprovisional Google 1,362 ±10 43 +16 A ›
134 C Solar Pro 4provisional Upstage 1,361 ±12 43 +47◆ A ›
135 C GPT-5 Miniprovisional OpenAI 1,361 ±11 42 +2 A ›
136 C MiniMax M2.5provisional MiniMax · Open weights 1,359 ±7 42 +9 A ›
137 C Mistral Medium 3provisional Mistral AI 1,359 ±9 42 -31 A ›
138 C Claude Sonnet 4provisional Anthropic 1,356 ±10 41 -62 A ›
139 C grok-3-mini-highprovisional xAI 1,353 ±12 41 +4 A ›
140 C gpt-oss-120bprovisional OpenAI · Open weights 1,345 ±10 40 -2 A ›
141 C Nova 2 Liteprovisional Amazon 1,344 ±13 40 -36 A ›
142 C Trinity Large Thinkingprovisional Arcee AI · Open weights 1,341 ±9 40 -30 A ›
143 C Qwen3 Next 80B A3B Thinkingprovisional Alibaba · Open weights 1,340 ±15 39 -10 A ›
144 C intellect-3provisional Unknown 1,339 ±20 39 — A ›
145 C o4 Miniprovisional OpenAI 1,338 ±8 38 -37 A ›
146 C GLM 4.6Vprovisional Zhipu AI · Open weights 1,338 ±24 38 -12 A ›
147 C Nemotron 3 Superprovisional NVIDIA · Open weights 1,338 ±15 37 +0 A ›
148 C granite-4.2-30bprovisional IBM 1,338 ±25 37 — A ›
149 C GLM 4.7 Flashprovisional Zhipu AI · Open weights 1,337 ±12 36 +1 A ›
150 C MiniMax M1provisional MiniMax · Open weights 1,334 ±10 36 -36 A ›
151 C MiniMax M2provisional MiniMax · Open weights 1,332 ±21 36 -32 A ›
152 C step-3provisional StepFun 1,332 ±21 35 -17 A ›
153 C trinity-large-previewprovisional Unknown 1,330 ±8 35 — A ›
154 C step-1o-turbo-202506provisional StepFun 1,330 ±18 34 — A ›
155 C GPT-4.1 Miniprovisional OpenAI 1,330 ±9 34 -42 A ›
156 C Mistral Small 3.2provisional Mistral AI · Open weights 1,328 ±13 34 -14 A ›
157 C Step 2 (16K)provisional StepFun 1,327 ±18 33 -134 A ›
158 C gemini-1.5-pro-002provisional Google 1,327 ±6 33 — A ›
159 C ling-flash-2.0provisional Ant Group 1,314 ±21 32 — A ›
160 C Nemotron 3 Nano 30B A3Bprovisional NVIDIA · Open weights 1,312 ±11 32 +6 A ›
161 C Mercury 2provisional Inception Labs 1,308 ±22 31 -20 A ›
162 C GLM 4.5Vprovisional Zhipu AI · Open weights 1,305 ±22 31 -60 A ›
163 C GPT-5 Nanoprovisional OpenAI 1,304 ±19 30 -7 A ›
164 C Qwen3 30B A3Bprovisional Alibaba · Open weights 1,302 ±10 30 -29 A ›
165 C deepseek-v2.5-1210provisional DeepSeek 1,300 ±14 30 — A ›
166 C gemini-1.5-flash-002provisional Google 1,299 ±7 29 — A ›
167 C gemma-3n-e4b-itprovisional Google · Open weights 1,299 ±12 29 — A ›
168 C gemini-advanced-0514provisional Google 1,295 ±9 28 — A ›
169 C athene-v2-chatprovisional Unknown 1,294 ±8 28 — A ›
170 C Llama 3.1 Nemotron Ultra 253Bprovisional NVIDIA · Open weights 1,292 ±24 28 — A ›
171 C gpt-oss-20bprovisional OpenAI · Open weights 1,290 ±17 27 -26 A ›
172 C Granite 4.2 8Bprovisional IBM · Open weights 1,289 ±27 27 -14 A ›
173 C qwen2.5-plus-1127provisional Alibaba 1,284 ±13 26 — A ›
174 C olmo-3.1-32b-instructprovisional Ai2 1,282 ±12 26 — A ›
175 C GPT-4.1 Nanoprovisional OpenAI 1,276 ±17 25 -57 A ›
176 D nvidia-nemotron-3.5-lightning-30b-a3b-nvfp4provisional NVIDIA 1,274 ±13 25 +3 A ›
177 D yi-lightningprovisional Unknown 1,272 ±8 24 — A ›
178 D olmo-3-32b-thinkprovisional Ai2 1,270 ±20 24 — A ›
179 D athene-70b-0725provisional Unknown 1,268 ±11 24 — A ›
180 D granite-4.2-3bprovisional IBM 1,264 ±27 23 — A ›
181 D llama-3.1-tulu-3-70bprovisional Ai2 1,262 ±22 23 — A ›
182 D ring-flash-2.0provisional Ant Group 1,260 ±22 22 — A ›
183 D hunyuan-large-visionprovisional Tencent 1,260 ±22 22 — A ›
184 D amazon-nova-pro-v1.0provisional Amazon 1,259 ±9 22 — A ›
185 D gemini-1.5-flash-8b-001provisional Google 1,256 ±7 21 — A ›
186 D granite-4.1-8bprovisional IBM · Open weights 1,255 ±24 21 -28 A ›
187 D Llama 3.1 Nemotron 70B Instructprovisional NVIDIA · Open weights 1,248 ±14 20 -95 A ›
188 D gemma-2-9b-it-simpoprovisional Google 1,246 ±14 20 — A ›
189 D nemotron-4-340b-instructprovisional NVIDIA 1,242 ±11 19 — A ›
190 D amazon-nova-lite-v1.0provisional Amazon 1,239 ±10 19 — A ›
191 D glm-4-0520provisional Zhipu AI 1,236 ±13 18 — A ›
192 D command-r-plusprovisional Cohere 1,230 ±8 18 -220 A ›
193 D ibm-granite-h-smallprovisional IBM 1,223 ±23 18 — A ›
194 D jamba-1.5-largeprovisional AI21 Labs 1,222 ±15 17 -178 A ›
195 D Ministral 3 8B 2512provisional Mistral AI · Open weights 1,221 ±16 17 -133 A ›
196 D llama-3.1-tulu-3-8bprovisional Ai2 1,219 ±20 16 — A ›
197 D deepseek-coder-v2provisional DeepSeek 1,215 ±12 16 — A ›
198 D olmo-2-0325-32b-instructprovisional Ai2 1,214 ±26 16 — A ›
199 D llama-3.1-nemotron-51b-instructprovisional NVIDIA 1,214 ±18 15 — A ›
200 D gemini-pro-dev-apiprovisional Google 1,214 ±17 15 — A ›
201 D amazon-nova-micro-v1.0provisional Amazon 1,213 ±10 14 — A ›
202 D reka-flash-21b-20240226-onlineprovisional Unknown 1,195 ±13 14 — A ›
203 D hunyuan-standard-256kprovisional Tencent 1,191 ±21 13 — A ›
204 D mixtral-8x22b-instruct-v0.1provisional Mistral AI · Open weights 1,190 ±9 13 -236 A ›
205 D jamba-1.5-miniprovisional AI21 Labs 1,189 ±15 12 -160 A ›
206 D phi-3-medium-4k-instructprovisional Microsoft 1,179 ±10 12 -199 A ›
207 D qwen1.5-110b-chatprovisional Alibaba 1,157 ±10 12 — A ›
208 D zephyr-orpo-141b-A35b-v0.1provisional Unknown 1,156 ±22 11 — A ›
209 D phi-3-small-8k-instructprovisional Microsoft 1,150 ±11 11 -219 A ›
210 D internlm2_5-20b-chatprovisional Unknown 1,148 ±13 10 — A ›
211 D qwen1.5-72b-chatprovisional Alibaba 1,145 ±11 10 — A ›
212 D snowflake-arctic-instructprovisional Unknown 1,144 ±11 9 — A ›
213 D starling-lm-7b-betaprovisional Unknown 1,144 ±13 9 — A ›
214 D starling-lm-7b-alphaprovisional Unknown 1,140 ±22 9 — A ›
215 D granite-3.1-8b-instructprovisional IBM 1,137 ±23 8 — A ›
216 D yi-1.5-34b-chatprovisional Unknown 1,134 ±10 8 — A ›
217 D vicuna-33bprovisional Unknown 1,128 ±18 7 — A ›
218 D dbrx-instruct-previewprovisional Unknown 1,123 ±11 7 — A ›
219 D qwen1.5-32b-chatprovisional Alibaba 1,119 ±11 6 — A ›
220 D granite-3.1-2b-instructprovisional IBM 1,114 ±23 6 — A ›
221 D granite-3.0-8b-instructprovisional IBM 1,109 ±16 6 — A ›
222 D yi-34b-chatprovisional Unknown 1,101 ±17 5 — A ›
223 D granite-3.0-2b-instructprovisional IBM 1,097 ±16 5 — A ›
224 D qwen1.5-14b-chatprovisional Alibaba 1,097 ±13 4 — A ›
225 D vicuna-13bprovisional Unknown 1,096 ±22 4 — A ›
226 D gemma-1.1-7b-itprovisional Google 1,096 ±11 3 — A ›
227 D phi-3-mini-4k-instructprovisional Microsoft 1,076 ±11 3 -243 A ›
228 D phi-3-mini-4k-instruct-june-2024provisional Microsoft 1,075 ±14 3 — A ›
229 D mistral-7b-instruct-v0.2provisional Mistral AI 1,074 ±14 2 -230 A ›
230 D phi-3-mini-128k-instructprovisional Microsoft 1,062 ±13 2 -272 A ›
231 D gemma-1.1-2b-itprovisional Google 1,051 ±15 1 — A ›
232 D smollm2-1.7b-instructprovisional Unknown 1,049 ±25 1 — A ›
233 D qwen1.5-4b-chatprovisional Alibaba 1,022 ±20 0 — A ›
234 D olmo-7b-instructprovisional Ai2 1,019 ±23 0 — A ›
…