Rankings based on user votes and Elo ratings
# | Model(18/18) | Elo | Win % | Votes | Record |
|---|---|---|---|---|---|
| 1 | GPT-5.2 | 1267 | 52.3% | 172 | 90W-55L-27T |
| 2 | MiniMax M3 | 1265 | 52.8% | 53 | 28W-21L-4T |
| 3 | Kimi K2.6 | 1264 | 58.1% | 31 | 18W-12L-1T |
| 4 | GPT-5.6 Luna | 1243 | 42.9% | 42 | 18W-23L-1T |
| 5 | Claude Haiku 4.5 | 1241 | 39.5% | 185 | 73W-83L-29T |
| 6 | Mistral Medium 3.5 | 1219 | 41.9% | 31 | 13W-14L-4T |
| 7 | Functionary Swahili Large | 1216 | 47.1% | 157 | 74W-59L-24T |
| 8 | GLM 5.2 | 1208 | 53.1% | 49 | 26W-20L-3T |
| 9 | GLM 5 | 1195 | 43.7% | 199 | 87W-83L-29T |
| 10 | Functionary Swahili Mini | 1191 | 49.4% | 77 | 38W-33L-6T |
| 11 | Claude Sonnet 4.6 | 1190 | 42.0% | 176 | 74W-82L-20T |
| 12 | Claude Sonnet 4.5 | 1190 | 42.3% | 168 | 71W-68L-29T |
| 13 | Qwen3.7 Max | 1164 | 45.5% | 33 | 15W-18L |
| 14 | GPT-5 Nano | 1159 | 32.9% | 164 | 54W-88L-22T |
| 15 | GPT-oss-120B | 1122 | 40.2% | 164 | 66W-72L-26T |
| 16 | DeepSeek V4 Flash | 1118 | 34.1% | 44 | 15W-29L |
| 17 | DeepSeek V4 Pro | 1229 | 44.8% | 29 | 13W-13L-3T |
| 18 | Step 3.7 Flash | 1192 | 44.4% | 18 | 8W-9L-1T |
Click column headers to sort by category
Model(18/18) | Overall | Long Prompt | Hard Prompt | Math | Instruction-Following | Coding |
|---|---|---|---|---|---|---|
| GPT-5.2 | 1 | — | 4 | 6 | 1 | 9 |
| MiniMax M3 | 2 | — | 11 | — | — | — |
| Kimi K2.6 | 3 | — | 10 | — | — | — |
| GPT-5.6 Luna | 4 | — | 9 | — | — | — |
| Claude Haiku 4.5 | 5 | — | 1 | 5 | 4 | 1 |
| Mistral Medium 3.5 | 6 | — | — | — | — | — |
| Functionary Swahili Large | 7 | — | 6 | — | 10 | 2 |
| GLM 5.2 | 8 | — | 17 | — | — | — |
| GLM 5 | 9 | — | 2 | 2 | 2 | 4 |
| Functionary Swahili Mini | 10 | — | 5 | — | 7 | 3 |
| Claude Sonnet 4.6 | 11 | — | 7 | 4 | 9 | 7 |
| Claude Sonnet 4.5 | 12 | — | 8 | 3 | 5 | 5 |
| Qwen3.7 Max | 13 | — | 15 | — | 6 | — |
| GPT-5 Nano | 14 | 2 | 3 | 7 | 11 | 8 |
| GPT-oss-120B | 15 | 1 | 16 | 1 | 8 | 6 |
| DeepSeek V4 Flash | 16 | — | 12 | — | — | — |
| DeepSeek V4 Pro | 17 | — | 14 | — | 3 | — |
| Step 3.7 Flash | 18 | — | 13 | — | — | — |