AI中转站/模型比价

模型能力排名

基于 OpenRouter Benchmarks数据，按各能力分数从高到低展示，数值越大代表能力越优秀。

综合智能指数

指数 从高到低
排名模型供应商上下文指数
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 59.9
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 55.7
3 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 54.8
4 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 53.5
5 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 53.4
6 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 51.4
7 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 51.1
8 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 50.2
9 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 47.2
10 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 46.5
11 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 46
12 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 44.4
13 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 44.3
14 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 42.8
15 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 42.2
16 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 41.9
17 DeepSeek: DeepSeek V4 Flashdeepseek/deepseek-v4-flash deepseek 1M 40.3
18 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 40.2
19 OpenAI: GPT-5.4 Miniopenai/gpt-5.4-mini openai 400K 40
20 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 39.6
21 Qwen: Qwen3.7 Plusqwen/qwen3.7-plus qwen 1M 39
22 OpenAI: GPT-5.4 Nanoopenai/gpt-5.4-nano openai 400K 38.2
23 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 38.1
24 NVIDIA: Nemotron 3 Ultranvidia/nemotron-3-ultra-550b-a55b nvidia 1M 37.8
25 NVIDIA: Nemotron 3 Ultra (free)nvidia/nemotron-3-ultra-550b-a55b:free nvidia 1M 37.8
26 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 37.6
27 Qwen: Qwen3.6 27Bqwen/qwen3.6-27b qwen 262.1K 37.1
28 OpenAI: GPT-5.1openai/gpt-5.1 openai 400K 36.9
29 Anthropic: Claude Sonnet 4.5anthropic/claude-sonnet-4.5 anthropic 1M 36.4
30 OpenAI: GPT-5openai/gpt-5 openai 400K 34.7

编码能力指数

指数 从高到低
排名模型供应商上下文指数
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 76.5
2 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 74.9
3 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 74.3
4 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 73.6
5 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 71.5
6 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 71.1
7 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 70.1
8 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 68.8
9 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 68.8
10 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 66
11 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 63
12 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 60.8
13 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 60.2
14 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 59.4
15 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 58.6
16 DeepSeek: DeepSeek V4 Flashdeepseek/deepseek-v4-flash deepseek 1M 56.2
17 OpenAI: GPT-5.4 Miniopenai/gpt-5.4-mini openai 400K 56.1
18 OpenAI: GPT-5.4 Nanoopenai/gpt-5.4-nano openai 400K 56.1
19 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 56
20 Qwen: Qwen3.7 Plusqwen/qwen3.7-plus qwen 1M 55.9
21 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 55.8
22 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 54.5
23 Qwen: Qwen3.6 27Bqwen/qwen3.6-27b qwen 262.1K 53.7
24 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 52.6
25 Anthropic: Claude Sonnet 4.5anthropic/claude-sonnet-4.5 anthropic 1M 52.1
26 OpenAI: GPT-5.1openai/gpt-5.1 openai 400K 49.4
27 NVIDIA: Nemotron 3 Ultranvidia/nemotron-3-ultra-550b-a55b nvidia 1M 49.3
28 NVIDIA: Nemotron 3 Ultra (free)nvidia/nemotron-3-ultra-550b-a55b:free nvidia 1M 49.3
29 Qwen: Qwen3.5 397B A17Bqwen/qwen3.5-397b-a17b qwen 256K 48.2
30 Mistral: Mistral Medium 3.5mistralai/mistral-medium-3-5 mistralai 262.1K 46.9

智能体能力指数

指数 从高到低
排名模型供应商上下文指数
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 52.8
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 47.2
3 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 46.7
4 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 44.9
5 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 44.4
6 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 43.1
7 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 41.1
8 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 40.8
9 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 37.4
10 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 36.4
11 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 35.4
12 DeepSeek: DeepSeek V4 Flashdeepseek/deepseek-v4-flash deepseek 1M 31.1
13 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 30.6
14 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 30.3
15 OpenAI: GPT-5.4 Miniopenai/gpt-5.4-mini openai 400K 30.2
16 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 29.9
17 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 29.6
18 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 29.1
19 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 27.6
20 OpenAI: GPT-5.4 Nanoopenai/gpt-5.4-nano openai 400K 27.5
21 NVIDIA: Nemotron 3 Ultranvidia/nemotron-3-ultra-550b-a55b nvidia 1M 27.4
22 NVIDIA: Nemotron 3 Ultra (free)nvidia/nemotron-3-ultra-550b-a55b:free nvidia 1M 27.4
23 Qwen: Qwen3.6 27Bqwen/qwen3.6-27b qwen 262.1K 27
24 OpenAI: GPT-5openai/gpt-5 openai 400K 25.7
25 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 25.6
26 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 25.4
27 Anthropic: Claude Sonnet 4.5anthropic/claude-sonnet-4.5 anthropic 1M 24.6
28 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 24.1
29 StepFun: Step 3.7 Flashstepfun/step-3.7-flash stepfun 256K 21.5
30 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 21.4

Agent 场景 / agenticgamedev

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1280 63.4% 1
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1257 62% 2
3 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1211 55% 3
4 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1210 54% 4
5 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1206 53% 5
6 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1195 50.9% 6
7 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1180 50.3% 7
8 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1166 48.2% 8
9 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1144 43.7% 9
10 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1127 41.2% 10
11 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1028 27.5% 11

Agent 场景 / agentichtmlslides

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1254 59.4% 1
2 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1248 59% 2
3 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1243 58% 3
4 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1227 55.6% 4
5 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1226 55.8% 5
6 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1205 52% 6
7 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1162 45.8% 7
8 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1134 41.7% 8
9 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1084 34.2% 9
10 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1066 31.8% 10

Agent 场景 / agenticslides

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1334 64.7% 1
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1294 64.8% 2
3 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1245 54.4% 3
4 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1244 57.5% 4
5 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1187 45.8% 5
6 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1171 51.6% 6
7 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1150 43.5% 7
8 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1112 33.8% 8
9 Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview google 1M 1073 39.3% 9
10 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1072 31.9% 10

Agent 场景 / agenticslides(html)

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1252 59.5% 1
2 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1252 59.2% 2
3 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1242 57.8% 3
4 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1230 56% 4
5 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1219 54.4% 5
6 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1204 51.8% 6
7 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1162 45.7% 7
8 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1137 41.8% 8
9 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1077 33.2% 9
10 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1066 31.7% 10

Agent 场景 / agenticslides(python-pptx)

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1335 66.5% 1
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1310 68.9% 2
3 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1242 57.8% 3
4 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1240 53.3% 4
5 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1186 45.5% 5
6 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1183 53.9% 6
7 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1155 45.2% 7
8 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1107 33.9% 8
9 Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview google 1M 1075 39.3% 9
10 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1068 32.4% 10

Agent 场景 / Android 原生

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1350 68.2% 1
2 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1325 61.7% 2
3 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1267 54.8% 3
4 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1261 55.3% 4
5 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1256 54.4% 5
6 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1256 57.5% 6
7 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1247 53.9% 7
8 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1243 61.7% 8
9 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1234 61.9% 9
10 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1212 68.8% 10
11 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1211 49.8% 11
12 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1210 50.1% 12
13 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1186 65.5% 13
14 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1186 46.3% 14
15 OpenAI: GPT-5.2-Codexopenai/gpt-5.2-codex openai 400K 1176 47.5% 15
16 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1175 44.1% 16
17 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1129 57.9% 17
18 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 1123 56% 18
19 Z.ai: GLM 4.6z-ai/glm-4.6 z-ai 202.8K 1093 52.6% 20
20 OpenAI: GPT-5.3-Codexopenai/gpt-5.3-codex openai 400K 1084 35.2% 21
21 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1078 28.9% 22
22 OpenAI: GPT-5.2openai/gpt-5.2 openai 400K 1069 49.2% 23
23 Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview google 1M 1058 48% 24
24 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1054 39% 25
25 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1053 26.7% 26
26 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1036 47.4% 27

Agent 场景 / godotgamedev

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1277 60.6% 1
2 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1254 59.5% 2
3 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1237 54.8% 3
4 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1236 55.9% 4
5 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1221 53.6% 5
6 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1219 53.5% 6
7 Z.ai: GLM 4.6z-ai/glm-4.6 z-ai 202.8K 1219 53.2% 7
8 Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview google 1M 1218 52.3% 8
9 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1217 53% 9
10 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1214 53% 10
11 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1210 50.8% 11
12 OpenAI: GPT-5.2-Codexopenai/gpt-5.2-codex openai 400K 1187 48% 12
13 OpenAI: GPT-5.2openai/gpt-5.2 openai 400K 1182 48.1% 13
14 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1172 46.9% 14
15 OpenAI: GPT-5.3-Codexopenai/gpt-5.3-codex openai 400K 1162 45.1% 15
16 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1135 38.5% 16
17 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1134 38.6% 17
18 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 1131 39.1% 18
19 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 1097 34% 20

Agent 场景 / HTML 幻灯片

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1260 60.8% 1
2 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1247 58.6% 2
3 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1237 57.5% 3
4 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1222 55.6% 4
5 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1206 52.5% 5
6 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1205 52.4% 6
7 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1204 52.5% 7
8 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1203 52.4% 8
9 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1182 49% 9
10 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1136 42% 10
11 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1084 34.4% 11
12 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1065 31.8% 12

Agent 场景 / pptxslides

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1344 67.3% 1
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1306 67.9% 2
3 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1244 57.7% 3
4 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1241 53.5% 4
5 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1181 44.3% 5
6 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1164 52.1% 6
7 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1157 45.3% 7
8 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1110 34.1% 8
9 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1071 32.5% 9

Agent 场景 / python-pptxslides

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1351 66.3% 1
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1298 65.9% 2
3 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1258 54.2% 3
4 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1247 57.4% 4
5 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1219 50.4% 5
6 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1192 45.8% 6
7 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1180 42.1% 7
8 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1169 51.7% 8
9 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1152 43.3% 9
10 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1109 31.9% 10
11 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1072 30.8% 11

Agent 场景 / Web 应用

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1330 65.5% 1
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1288 55.2% 2
3 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1284 57.5% 3
4 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1271 52.2% 4
5 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1268 60.9% 5
6 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1264 54.2% 6
7 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1260 58.1% 7
8 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1247 55.8% 8
9 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1246 50.5% 9
10 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1245 55.1% 10
11 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1222 54.2% 11
12 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1214 50% 12
13 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1197 48.1% 13
14 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1195 46.4% 14
15 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1195 50.3% 15
16 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1191 46.3% 16
17 Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview google 1M 1186 49.5% 17
18 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1183 45.2% 18
19 OpenAI: GPT-5.2openai/gpt-5.2 openai 400K 1158 45.7% 19
20 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1139 41.7% 20
21 Anthropic: Claude Sonnet 4.5anthropic/claude-sonnet-4.5 anthropic 1M 1131 43.1% 21
22 OpenAI: GPT-5.2-Codexopenai/gpt-5.2-codex openai 400K 1127 40.5% 22
23 OpenAI: GPT-5.3-Codexopenai/gpt-5.3-codex openai 400K 1119 39.1% 23
24 OpenAI: GPT-5.1-Codexopenai/gpt-5.1-codex openai 400K 1088 44.1% 24
25 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 1018 26.6% 25

Agent 场景 / 全栈应用

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1503 80.1% 1
2 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1321 65.7% 2
3 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1293 63.6% 3
4 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1286 68.6% 4
5 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1271 63.1% 5
6 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1264 58% 6
7 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1236 54.8% 7
8 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1231 56.8% 8
9 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1222 59.9% 9
10 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1221 48.1% 10
11 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1220 56.1% 11
12 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1212 54.4% 12
13 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1189 52.8% 13
14 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1180 54.2% 14
15 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1151 44.4% 15
16 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1138 44.9% 16
17 Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview google 1M 1128 47.1% 17
18 Anthropic: Claude Sonnet 4.5anthropic/claude-sonnet-4.5 anthropic 1M 1118 43.5% 18
19 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1117 41.1% 19
20 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 1115 44.9% 20
21 OpenAI: GPT-5.2openai/gpt-5.2 openai 400K 1108 44% 21
22 Z.ai: GLM 4.6z-ai/glm-4.6 z-ai 202.8K 1097 42.3% 22
23 OpenAI: GPT-5.1-Codexopenai/gpt-5.1-codex openai 400K 1090 44.5% 23
24 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1083 40.8% 24
25 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1075 31.8% 26
26 OpenAI: GPT-5.2-Codexopenai/gpt-5.2-codex openai 400K 1058 37% 27
27 OpenAI: GPT-5.3-Codexopenai/gpt-5.3-codex openai 400K 1049 36.4% 28
28 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 948 22.1% 29

Agent 场景 / 移动应用

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1283 59.4% 1
2 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1257 60.5% 2
3 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1257 56.5% 3
4 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1253 59.2% 4
5 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1240 54% 5
6 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1236 57.5% 6
7 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1231 55.1% 7
8 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1226 53.5% 8
9 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1220 51% 9
10 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1220 50.3% 10
11 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1218 53% 11
12 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1217 52.3% 12
13 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1209 47.3% 13
14 OpenAI: GPT-5.1-Codexopenai/gpt-5.1-codex openai 400K 1199 54% 14
15 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1194 50.9% 15
16 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 1185 49.9% 16
17 Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview google 1M 1183 49.8% 17
18 Z.ai: GLM 4.6z-ai/glm-4.6 z-ai 202.8K 1182 49.3% 18
19 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1182 48.4% 19
20 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1182 49.3% 20
21 Anthropic: Claude Sonnet 4.5anthropic/claude-sonnet-4.5 anthropic 1M 1180 48.9% 21
22 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1172 46.5% 22
23 OpenAI: GPT-5.2openai/gpt-5.2 openai 400K 1169 47.4% 23
24 OpenAI: GPT-5.2-Codexopenai/gpt-5.2-codex openai 400K 1168 47.7% 24
25 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1149 44.4% 25
26 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1139 38.6% 26
27 OpenAI: GPT-5.3-Codexopenai/gpt-5.3-codex openai 400K 1129 41.6% 27
28 OpenAI: GPT-5 Codexopenai/gpt-5-codex openai 400K 1127 43.2% 28
29 OpenAI: GPT-5.1openai/gpt-5.1 openai 400K 1123 43.8% 30

模型场景 / 3D

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1373 64% 1
2 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1368 67.2% 2
3 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1355 62.9% 3
4 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1352 65.2% 4
5 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1340 63.2% 5
6 Z.ai: GLM 5 Turboz-ai/glm-5-turbo z-ai 262.1K 1332 60.6% 6
7 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 1325 60.6% 7
8 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1325 59.7% 8
9 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1322 60.2% 9
10 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 1322 59.1% 10
11 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 1321 58% 11
12 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1316 60.4% 12
13 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1315 57.3% 13
14 Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro nex-agi 262.1K 1314 55.9% 14
15 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1313 60.2% 15
16 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1308 56.8% 16
17 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1305 56.6% 17
18 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1304 61.3% 18
19 Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5 xiaomi 1M 1298 54.1% 19
20 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1290 59.7% 20
21 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1289 56.3% 21
22 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1289 55.1% 22
23 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1286 54% 23
24 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 1273 51.5% 24
25 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 1269 54.4% 25
26 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 1268 51% 26
27 DeepSeek: DeepSeek V4 Flashdeepseek/deepseek-v4-flash deepseek 1M 1268 50% 27
28 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1266 53.8% 28
29 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1265 53.2% 29
30 Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview google 1M 1261 62.7% 30

模型场景 / asciiart

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1369 70.1% 1
2 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1328 66.6% 2
3 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1314 62.7% 3
4 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1312 63.3% 4
5 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1311 62.2% 5
6 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1305 62.8% 6
7 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1304 63.2% 7
8 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1278 60.5% 8
9 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1252 53.2% 9
10 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1245 55.3% 10
11 Anthropic: Claude Sonnet 4.5anthropic/claude-sonnet-4.5 anthropic 1M 1241 55.6% 11
12 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1235 54.9% 12
13 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1225 49.8% 13
14 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1221 49.8% 14
15 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 1220 50.6% 15
16 StepFun: Step 3.7 Flashstepfun/step-3.7-flash stepfun 256K 1219 50.7% 16
17 Google: Gemini 3.1 Flash Lite Previewgoogle/gemini-3.1-flash-lite-preview google 1M 1213 50.7% 17
18 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 1212 48.3% 18
19 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1211 46.9% 19
20 Anthropic: Claude Opus 4.1anthropic/claude-opus-4.1 anthropic 200K 1210 51.4% 20
21 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 1202 47.6% 21
22 OpenAI: GPT-5.2openai/gpt-5.2 openai 400K 1199 50.7% 22
23 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 1199 50.3% 23
24 Z.ai: GLM 5 Turboz-ai/glm-5-turbo z-ai 262.1K 1197 50.5% 24
25 OpenAI: GPT-5.3-Codexopenai/gpt-5.3-codex openai 400K 1197 51.2% 25
26 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1194 47.9% 26
27 xAI: Grok 4.3x-ai/grok-4.3 x-ai 1M 1193 47.5% 27
28 Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5 xiaomi 1M 1191 47.5% 28
29 OpenAI: GPT-5openai/gpt-5 openai 400K 1187 49% 29
30 Anthropic: Claude Haiku 4.5anthropic/claude-haiku-4.5 anthropic 200K 1186 49.3% 30

模型场景 / graphicdesign

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)google/gemini-3.1-flash-image-preview google 131.1K 1295 66.5% 2
2 Google: Nano Banana Pro (Gemini 3 Pro Image Preview)google/gemini-3-pro-image-preview google 65.5K 1286 66.1% 3
3 OpenAI: GPT-5 Imageopenai/gpt-5-image openai 400K 1210 51.4% 6
4 Google: Nano Banana (Gemini 2.5 Flash Image)google/gemini-2.5-flash-image google 32.8K 1203 56.9% 8
5 OpenAI: GPT-5 Image Miniopenai/gpt-5-image-mini openai 400K 1203 48.6% 9

模型场景 / image

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)google/gemini-3.1-flash-image-preview google 131.1K 1304 65.3% 2
2 Google: Nano Banana Pro (Gemini 3 Pro Image Preview)google/gemini-3-pro-image-preview google 65.5K 1272 62.2% 3
3 OpenAI: GPT-5 Imageopenai/gpt-5-image openai 400K 1219 54.1% 7
4 Google: Nano Banana (Gemini 2.5 Flash Image)google/gemini-2.5-flash-image google 32.8K 1214 55.6% 8
5 OpenAI: GPT-5 Image Miniopenai/gpt-5-image-mini openai 400K 1212 51.1% 9

模型场景 / imageediting

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Google: Nano Banana Pro (Gemini 3 Pro Image Preview)google/gemini-3-pro-image-preview google 65.5K 1273 66.1% 2

模型场景 / logo

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)google/gemini-3.1-flash-image-preview google 131.1K 1282 63.2% 2
2 Google: Nano Banana Pro (Gemini 3 Pro Image Preview)google/gemini-3-pro-image-preview google 65.5K 1260 61% 3
3 OpenAI: GPT-5 Image Miniopenai/gpt-5-image-mini openai 400K 1229 52.6% 5
4 OpenAI: GPT-5 Imageopenai/gpt-5-image openai 400K 1222 53.7% 6
5 Google: Nano Banana (Gemini 2.5 Flash Image)google/gemini-2.5-flash-image google 32.8K 1189 51.4% 9

模型场景 / svg

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1370 71.8% 1
2 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1347 70.3% 2
3 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1307 63% 3
4 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1287 62.6% 4
5 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1283 59.9% 5
6 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1282 61.6% 6
7 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1282 61.5% 7
8 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1276 60.1% 8
9 Z.ai: GLM 5 Turboz-ai/glm-5-turbo z-ai 262.1K 1270 59.5% 9
10 Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro nex-agi 262.1K 1267 55.3% 10
11 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1256 59.6% 11
12 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1254 52.5% 12
13 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1253 57.9% 13
14 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 1251 54.3% 14
15 OpenAI: GPT-5openai/gpt-5 openai 400K 1246 64.1% 15
16 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1243 54.3% 16
17 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 1241 55.5% 17
18 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1237 58.7% 18
19 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1235 53.9% 19
20 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1235 52.5% 20
21 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1225 54.4% 21
22 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1221 54.3% 22
23 Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5 xiaomi 1M 1220 52.4% 23
24 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 1219 52% 24
25 DeepSeek: DeepSeek V4 Flashdeepseek/deepseek-v4-flash deepseek 1M 1215 49.3% 25
26 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1211 50.1% 26
27 Anthropic: Claude Opus 4.1anthropic/claude-opus-4.1 anthropic 200K 1210 60.8% 27
28 MiniMax: MiniMax M2.5minimax/minimax-m2.5 minimax 204.8K 1208 54.5% 28
29 OpenAI: GPT-5.1openai/gpt-5.1 openai 400K 1207 57.4% 29
30 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1204 50.9% 30

模型场景 / UI 组件

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1414 71.3% 1
2 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1366 64.8% 2
3 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1351 64.4% 3
4 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1336 61.8% 4
5 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1335 59.1% 5
6 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1331 60.1% 6
7 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1324 61.7% 7
8 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 1322 57.2% 8
9 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1320 68.1% 9
10 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1320 57.2% 10
11 Z.ai: GLM 5 Turboz-ai/glm-5-turbo z-ai 262.1K 1319 58% 11
12 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1312 59.5% 12
13 Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5 xiaomi 1M 1310 56% 13
14 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1302 55.1% 14
15 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 1299 55.7% 15
16 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1298 56.9% 16
17 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1294 54.1% 17
18 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1292 57.6% 18
19 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1291 58.6% 19
20 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1289 54.3% 20
21 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1285 54% 21
22 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1283 54.8% 22
23 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 1282 52.2% 23
24 MiniMax: MiniMax M2.1minimax/minimax-m2.1 minimax 204.8K 1277 60.9% 24
25 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 1272 52% 25
26 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1269 51.9% 26
27 Z.ai: GLM 4.7 Flashz-ai/glm-4.7-flash z-ai 202.8K 1261 57.6% 27
28 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 1260 49.8% 28
29 Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro nex-agi 262.1K 1259 47.7% 29
30 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1256 52.7% 30

模型场景 / 代码分类

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1360 61.9% 1
2 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1349 63.3% 2
3 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1342 63.7% 3
4 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1336 60.9% 4
5 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1331 60% 5
6 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1327 58.1% 6
7 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1326 61.5% 7
8 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 1323 57.8% 8
9 Z.ai: GLM 5 Turboz-ai/glm-5-turbo z-ai 262.1K 1320 57.6% 9
10 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 1319 58.1% 10
11 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1311 57.4% 11
12 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1306 58.4% 12
13 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1306 55.4% 13
14 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1306 55.6% 14
15 Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5 xiaomi 1M 1303 55% 15
16 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1296 55.7% 16
17 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1296 56.7% 17
18 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1291 59.9% 18
19 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1290 64.2% 19
20 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 1287 54.8% 20
21 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1286 54.7% 21
22 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1282 54.8% 22
23 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 1281 53.3% 23
24 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1278 52.5% 24
25 Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro nex-agi 262.1K 1275 50.1% 25
26 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 1273 51.6% 26
27 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1268 54.8% 27
28 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 1266 54.9% 28
29 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1262 53.1% 29
30 MiniMax: MiniMax M2.5minimax/minimax-m2.5 minimax 204.8K 1256 56.8% 30

模型场景 / 数据可视化

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1381 70.1% 1
2 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1366 67% 2
3 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1326 59.9% 3
4 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1324 61.9% 4
5 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1322 61.4% 5
6 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1320 61.1% 6
7 Z.ai: GLM 5 Turboz-ai/glm-5-turbo z-ai 262.1K 1308 58.2% 7
8 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1302 56.7% 8
9 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1301 58.4% 9
10 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 1295 58.4% 10
11 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1291 56.7% 11
12 Google: Gemini 2.5 Progoogle/gemini-2.5-pro google 1M 1290 71.8% 12
13 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1288 55.7% 13
14 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1286 58.9% 14
15 OpenAI: GPT-5openai/gpt-5 openai 400K 1282 63.3% 15
16 Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5 xiaomi 1M 1282 54.1% 16
17 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1279 55.3% 17
18 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 1278 53.9% 18
19 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1276 56.6% 19
20 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 1272 53.3% 20
21 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1271 52.9% 21
22 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1270 53% 22
23 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1269 60.6% 23
24 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1266 54.8% 24
25 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 1262 50.7% 25
26 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1254 51% 26
27 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1252 53% 27
28 MiniMax: MiniMax M2.1minimax/minimax-m2.1 minimax 204.8K 1250 57% 28
29 Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro nex-agi 262.1K 1249 48.8% 29
30 OpenAI: GPT-5.1openai/gpt-5.1 openai 400K 1248 58% 30

模型场景 / 游戏开发

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1379 65.1% 1
2 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1357 61.3% 2
3 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1348 62.8% 3
4 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1344 63.2% 4
5 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 1344 59.8% 5
6 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1343 63.1% 6
7 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1338 61.9% 7
8 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 1338 61.1% 8
9 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1330 59.3% 9
10 Z.ai: GLM 5 Turboz-ai/glm-5-turbo z-ai 262.1K 1325 59.6% 10
11 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1323 58.7% 11
12 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1322 60.4% 12
13 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1316 58.1% 13
14 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1306 57.9% 14
15 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1301 54.9% 15
16 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1299 57.4% 16
17 Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5 xiaomi 1M 1298 55.6% 17
18 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 1296 56.1% 18
19 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1295 59.5% 19
20 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1288 51.6% 20
21 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1287 54.6% 21
22 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1281 51.5% 22
23 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 1273 53.3% 23
24 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1272 53.6% 24
25 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 1270 51.9% 25
26 Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro nex-agi 262.1K 1269 48.7% 26
27 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1264 54.5% 27
28 DeepSeek: DeepSeek V4 Flashdeepseek/deepseek-v4-flash deepseek 1M 1263 50.7% 28
29 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1262 53.9% 29
30 OpenAI: GPT-5.2openai/gpt-5.2 openai 400K 1260 56% 30

模型场景 / 网站

Elo 从高到低
排名模型供应商上下文Elo胜率源排名
1 Z.ai: GLM 5.2z-ai/glm-5.2 z-ai 1M 1356 61.6% 1
2 Anthropic: Claude Fable 5anthropic/claude-fable-5 anthropic 1M 1341 62.5% 2
3 Anthropic: Claude Opus 4.6anthropic/claude-opus-4.6 anthropic 1M 1337 63.1% 3
4 Anthropic: Claude Opus 4.7anthropic/claude-opus-4.7 anthropic 1M 1335 60.4% 4
5 Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.6 anthropic 1M 1327 61.6% 5
6 Anthropic: Claude Sonnet 5anthropic/claude-sonnet-5 anthropic 1M 1322 58.2% 6
7 Z.ai: GLM 5.1z-ai/glm-5.1 z-ai 202.8K 1318 57.2% 7
8 MoonshotAI: Kimi K2.6moonshotai/kimi-k2.6 moonshotai 262.1K 1318 56.4% 8
9 MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code moonshotai 262.1K 1317 57% 9
10 Z.ai: GLM 5 Turboz-ai/glm-5-turbo z-ai 262.1K 1315 56.5% 10
11 Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro xiaomi 1M 1310 56.6% 11
12 Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5 xiaomi 1M 1304 55.2% 12
13 MiniMax: MiniMax M3minimax/minimax-m3 minimax 1M 1304 55.6% 13
14 Qwen: Qwen3.7 Maxqwen/qwen3.7-max qwen 1M 1302 56.3% 14
15 Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash google 1M 1299 57.4% 15
16 Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview google 1M 1294 64.3% 16
17 Z.ai: GLM 5z-ai/glm-5 z-ai 202.8K 1291 55.1% 17
18 MoonshotAI: Kimi K2.5moonshotai/kimi-k2.5 moonshotai 262.1K 1291 55.9% 18
19 OpenAI: GPT-5.5openai/gpt-5.5 openai 1.1M 1290 55.9% 19
20 Anthropic: Claude Opus 4.5anthropic/claude-opus-4.5 anthropic 200K 1289 59.9% 20
21 MiniMax: MiniMax M2.7minimax/minimax-m2.7 minimax 204.8K 1287 54.3% 21
22 Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 anthropic 1M 1281 54.8% 22
23 DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro deepseek 1M 1278 53.6% 23
24 xAI: Grok 4.20x-ai/grok-4.20 x-ai 2M 1271 55.3% 24
25 Z.ai: GLM 5V Turboz-ai/glm-5v-turbo z-ai 202.8K 1270 51% 25
26 Z.ai: GLM 4.7z-ai/glm-4.7 z-ai 202.8K 1268 55.5% 26
27 MiniMax: MiniMax M2.5minimax/minimax-m2.5 minimax 204.8K 1264 57.5% 27
28 OpenAI: GPT-5.4openai/gpt-5.4 openai 1.1M 1264 53% 28
29 Qwen: Qwen3.6 Plusqwen/qwen3.6-plus qwen 1M 1262 51.4% 29
30 Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro nex-agi 262.1K 1258 48% 30