Update README.md
Browse files
README.md
CHANGED
|
@@ -55,134 +55,205 @@ license: mit
|
|
| 55 |
<table>
|
| 56 |
<thead>
|
| 57 |
<tr>
|
| 58 |
-
<th
|
| 59 |
-
<th align="center"><strong>Vision</strong></th>
|
| 60 |
-
<th align="center" colspan="
|
| 61 |
-
<th align="center" colspan="
|
| 62 |
</tr>
|
| 63 |
<tr>
|
| 64 |
<th></th>
|
| 65 |
<th></th>
|
| 66 |
-
<th align="center"
|
| 67 |
-
<th align="center"
|
| 68 |
-
<th align="center"
|
| 69 |
-
<th align="center"
|
| 70 |
-
<th align="center"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
</tr>
|
| 72 |
<tr>
|
| 73 |
<th></th>
|
| 74 |
<th></th>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
<th align="center">pass@1</th>
|
| 76 |
<th align="center">pass@1</th>
|
| 77 |
<th align="center">pass@1</th>
|
| 78 |
-
<th align="center"
|
| 79 |
-
<th align="center"
|
|
|
|
|
|
|
|
|
|
| 80 |
</tr>
|
| 81 |
</thead>
|
| 82 |
<tbody>
|
| 83 |
<tr>
|
| 84 |
-
<td>
|
| 85 |
-
<td align="center"
|
| 86 |
-
<td align="center">
|
| 87 |
-
<td align="center">
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
<td align="center">49.0</td>
|
| 89 |
-
<td align="center"
|
| 90 |
-
<td align="center"
|
| 91 |
</tr>
|
| 92 |
<tr>
|
| 93 |
-
<td>
|
| 94 |
-
<td align="center"
|
| 95 |
-
<td align="center">
|
| 96 |
-
<td align="center">
|
| 97 |
-
<td align="center">
|
| 98 |
-
<td align="center"
|
| 99 |
-
<td align="center"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
</tr>
|
| 101 |
<tr>
|
| 102 |
-
<td>Deepseek
|
| 103 |
<td align="center">β</td>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 104 |
<td align="center">97.3</td>
|
| 105 |
<td align="center">79.8</td>
|
| 106 |
<td align="center">71.5</td>
|
| 107 |
-
<td align="center"
|
| 108 |
-
<td align="center"
|
|
|
|
|
|
|
|
|
|
| 109 |
</tr>
|
| 110 |
<tr>
|
| 111 |
-
<td>
|
| 112 |
-
<td align="center"
|
| 113 |
-
<td align="center">
|
| 114 |
-
<td align="center">
|
| 115 |
-
<td align="center">
|
| 116 |
-
<td align="center"
|
| 117 |
-
<td align="center"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 118 |
</tr>
|
| 119 |
<tr>
|
| 120 |
-
<td>GPT
|
| 121 |
<td align="center">β
</td>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 122 |
<td align="center">74.6</td>
|
| 123 |
<td align="center">9.3</td>
|
| 124 |
<td align="center">49.9</td>
|
| 125 |
-
<td align="center">
|
| 126 |
-
<td align="center">
|
|
|
|
|
|
|
|
|
|
| 127 |
</tr>
|
| 128 |
<tr>
|
| 129 |
-
<td>
|
| 130 |
<td align="center">β
</td>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 131 |
<td align="center">96.2</td>
|
| 132 |
<td align="center">77.5</td>
|
| 133 |
-
<td align="center"
|
| 134 |
-
<td align="center">74.9</td>
|
| 135 |
<td align="center">70.0</td>
|
|
|
|
|
|
|
|
|
|
|
|
|
| 136 |
</tr>
|
| 137 |
<tr>
|
| 138 |
-
<td>Qwen2.5
|
| 139 |
<td align="center">β
</td>
|
| 140 |
-
<td align="center"
|
| 141 |
-
<td align="center"
|
| 142 |
-
<td align="center"
|
| 143 |
-
<td align="center"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 144 |
<td align="center">70.2</td>
|
|
|
|
|
|
|
|
|
|
|
|
|
| 145 |
</tr>
|
| 146 |
<tr>
|
| 147 |
-
<td>
|
| 148 |
-
<td align="center">β
</td>
|
| 149 |
-
<td align="center">-</td>
|
| 150 |
-
<td align="center">-</td>
|
| 151 |
-
<td align="center">-</td>
|
| 152 |
-
<td align="center">67.5</td>
|
| 153 |
-
<td align="center">56.8</td>
|
| 154 |
-
</tr>
|
| 155 |
-
<tr>
|
| 156 |
-
<td>InternVL2-Llama3-76B</td>
|
| 157 |
-
<td align="center">β
</td>
|
| 158 |
-
<td align="center">-</td>
|
| 159 |
-
<td align="center">-</td>
|
| 160 |
-
<td align="center">-</td>
|
| 161 |
-
<td align="center">65.5</td>
|
| 162 |
-
<td align="center">62.7</td>
|
| 163 |
-
</tr>
|
| 164 |
-
<tr>
|
| 165 |
-
<td>InternVL2.5-78B</td>
|
| 166 |
<td align="center">β
</td>
|
| 167 |
-
<td align="center"
|
| 168 |
-
<td align="center"
|
| 169 |
-
<td align="center"
|
| 170 |
-
<td align="center"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 171 |
<td align="center">70.1</td>
|
| 172 |
-
|
| 173 |
-
|
| 174 |
-
<td>
|
| 175 |
-
<td align="center"
|
| 176 |
-
<td align="center">94.0</td>
|
| 177 |
-
<td align="center">72.0</td>
|
| 178 |
-
<td align="center">61.6</td>
|
| 179 |
-
<td align="center">67.5</td>
|
| 180 |
-
<td align="center">69.0</td>
|
| 181 |
</tr>
|
| 182 |
</tbody>
|
| 183 |
</table>
|
| 184 |
|
| 185 |
|
|
|
|
| 186 |
---
|
| 187 |
|
| 188 |
|
|
|
|
| 55 |
<table>
|
| 56 |
<thead>
|
| 57 |
<tr>
|
| 58 |
+
<th>Model</th>
|
| 59 |
+
<th align="center"><strong>Supports Vision</strong></th>
|
| 60 |
+
<th align="center" colspan="8"><strong>TextΒ ReasoningΒ (%)</strong></th>
|
| 61 |
+
<th align="center" colspan="5"><strong>MultimodalΒ ReasoningΒ (%)</strong></th>
|
| 62 |
</tr>
|
| 63 |
<tr>
|
| 64 |
<th></th>
|
| 65 |
<th></th>
|
| 66 |
+
<th align="center">AIME24</th>
|
| 67 |
+
<th align="center">LiveCodebench</th>
|
| 68 |
+
<th align="center">liveBench</th>
|
| 69 |
+
<th align="center">IFEVAL</th>
|
| 70 |
+
<th align="center">BFCL</th>
|
| 71 |
+
<th align="center">MATHβ500</th>
|
| 72 |
+
<th align="center">AIMEΒ 2024</th>
|
| 73 |
+
<th align="center">GPQA</th>
|
| 74 |
+
<th align="center">MMMU(val)</th>
|
| 75 |
+
<th align="center">MathVista(mini)</th>
|
| 76 |
+
<th align="center">MathVision(mini)</th>
|
| 77 |
+
<th align="center">OlympiadBench</th>
|
| 78 |
+
<th align="center">mmmuβpro</th>
|
| 79 |
</tr>
|
| 80 |
<tr>
|
| 81 |
<th></th>
|
| 82 |
<th></th>
|
| 83 |
+
<th align="center">%</th>
|
| 84 |
+
<th align="center">%</th>
|
| 85 |
+
<th align="center">%</th>
|
| 86 |
+
<th align="center">%</th>
|
| 87 |
+
<th align="center">%</th>
|
| 88 |
<th align="center">pass@1</th>
|
| 89 |
<th align="center">pass@1</th>
|
| 90 |
<th align="center">pass@1</th>
|
| 91 |
+
<th align="center">%</th>
|
| 92 |
+
<th align="center">%</th>
|
| 93 |
+
<th align="center">%</th>
|
| 94 |
+
<th align="center">%</th>
|
| 95 |
+
<th align="center">%</th>
|
| 96 |
</tr>
|
| 97 |
</thead>
|
| 98 |
<tbody>
|
| 99 |
<tr>
|
| 100 |
+
<td>R1V2β38B</td>
|
| 101 |
+
<td align="center">β
</td>
|
| 102 |
+
<td align="center">78.9</td>
|
| 103 |
+
<td align="center">63.6</td>
|
| 104 |
+
<td align="center">73.2</td>
|
| 105 |
+
<td align="center">82.9</td>
|
| 106 |
+
<td align="center">66.3</td>
|
| 107 |
+
<td align="center">94.0</td>
|
| 108 |
+
<td align="center">72.0</td>
|
| 109 |
+
<td align="center">61.6</td>
|
| 110 |
+
<td align="center">73.6</td>
|
| 111 |
+
<td align="center">74.0</td>
|
| 112 |
<td align="center">49.0</td>
|
| 113 |
+
<td align="center">62.6</td>
|
| 114 |
+
<td align="center">52.0</td>
|
| 115 |
</tr>
|
| 116 |
<tr>
|
| 117 |
+
<td>R1V1β38B</td>
|
| 118 |
+
<td align="center">β
</td>
|
| 119 |
+
<td align="center">72.0</td>
|
| 120 |
+
<td align="center">57.2</td>
|
| 121 |
+
<td align="center">54.6</td>
|
| 122 |
+
<td align="center">72.5</td>
|
| 123 |
+
<td align="center">53.5</td>
|
| 124 |
+
<td align="center">β</td>
|
| 125 |
+
<td align="center">β</td>
|
| 126 |
+
<td align="center">β</td>
|
| 127 |
+
<td align="center">68.0</td>
|
| 128 |
+
<td align="center">67.0</td>
|
| 129 |
+
<td align="center">β</td>
|
| 130 |
+
<td align="center">40.4</td>
|
| 131 |
+
<td align="center">β</td>
|
| 132 |
</tr>
|
| 133 |
<tr>
|
| 134 |
+
<td>DeepseekβR1β671B</td>
|
| 135 |
<td align="center">β</td>
|
| 136 |
+
<td align="center">74.3</td>
|
| 137 |
+
<td align="center">65.9</td>
|
| 138 |
+
<td align="center">71.6</td>
|
| 139 |
+
<td align="center">83.3</td>
|
| 140 |
+
<td align="center">60.3</td>
|
| 141 |
<td align="center">97.3</td>
|
| 142 |
<td align="center">79.8</td>
|
| 143 |
<td align="center">71.5</td>
|
| 144 |
+
<td align="center">β</td>
|
| 145 |
+
<td align="center">β</td>
|
| 146 |
+
<td align="center">β</td>
|
| 147 |
+
<td align="center">β</td>
|
| 148 |
+
<td align="center">β</td>
|
| 149 |
</tr>
|
| 150 |
<tr>
|
| 151 |
+
<td>GPTβo1</td>
|
| 152 |
+
<td align="center">β</td>
|
| 153 |
+
<td align="center">79.8</td>
|
| 154 |
+
<td align="center">63.4</td>
|
| 155 |
+
<td align="center">72.2</td>
|
| 156 |
+
<td align="center">β</td>
|
| 157 |
+
<td align="center">β</td>
|
| 158 |
+
<td align="center">β</td>
|
| 159 |
+
<td align="center">β</td>
|
| 160 |
+
<td align="center">β</td>
|
| 161 |
+
<td align="center">β</td>
|
| 162 |
+
<td align="center">β</td>
|
| 163 |
+
<td align="center">β</td>
|
| 164 |
+
<td align="center">β</td>
|
| 165 |
+
<td align="center">β</td>
|
| 166 |
</tr>
|
| 167 |
<tr>
|
| 168 |
+
<td>GPTβo4βmini</td>
|
| 169 |
<td align="center">β
</td>
|
| 170 |
+
<td align="center">93.4</td>
|
| 171 |
+
<td align="center">74.6</td>
|
| 172 |
+
<td align="center">78.1</td>
|
| 173 |
+
<td align="center">β</td>
|
| 174 |
+
<td align="center">β</td>
|
| 175 |
<td align="center">74.6</td>
|
| 176 |
<td align="center">9.3</td>
|
| 177 |
<td align="center">49.9</td>
|
| 178 |
+
<td align="center">81.6</td>
|
| 179 |
+
<td align="center">84.3</td>
|
| 180 |
+
<td align="center">58.0</td>
|
| 181 |
+
<td align="center">β</td>
|
| 182 |
+
<td align="center">β</td>
|
| 183 |
</tr>
|
| 184 |
<tr>
|
| 185 |
+
<td>ClaudeΒ 3.5Β Sonnet</td>
|
| 186 |
<td align="center">β
</td>
|
| 187 |
+
<td align="center">β</td>
|
| 188 |
+
<td align="center">β</td>
|
| 189 |
+
<td align="center">β</td>
|
| 190 |
+
<td align="center">β</td>
|
| 191 |
+
<td align="center">β</td>
|
| 192 |
+
<td align="center">78.3</td>
|
| 193 |
+
<td align="center">16.0</td>
|
| 194 |
+
<td align="center">65.0</td>
|
| 195 |
+
<td align="center">66.4</td>
|
| 196 |
+
<td align="center">65.3</td>
|
| 197 |
+
<td align="center">β</td>
|
| 198 |
+
<td align="center">β</td>
|
| 199 |
+
<td align="center">β</td>
|
| 200 |
+
</tr>
|
| 201 |
+
<tr>
|
| 202 |
+
<td>KimiΒ k1.5Β longcot</td>
|
| 203 |
+
<td align="center">β
</td>
|
| 204 |
+
<td align="center">β</td>
|
| 205 |
+
<td align="center">β</td>
|
| 206 |
+
<td align="center">β</td>
|
| 207 |
+
<td align="center">β</td>
|
| 208 |
+
<td align="center">β</td>
|
| 209 |
<td align="center">96.2</td>
|
| 210 |
<td align="center">77.5</td>
|
| 211 |
+
<td align="center">β</td>
|
|
|
|
| 212 |
<td align="center">70.0</td>
|
| 213 |
+
<td align="center">74.9</td>
|
| 214 |
+
<td align="center">β</td>
|
| 215 |
+
<td align="center">β</td>
|
| 216 |
+
<td align="center">β</td>
|
| 217 |
</tr>
|
| 218 |
<tr>
|
| 219 |
+
<td>Qwen2.5βVLβ72BβInstruct</td>
|
| 220 |
<td align="center">β
</td>
|
| 221 |
+
<td align="center">β</td>
|
| 222 |
+
<td align="center">β</td>
|
| 223 |
+
<td align="center">β</td>
|
| 224 |
+
<td align="center">β</td>
|
| 225 |
+
<td align="center">β</td>
|
| 226 |
+
<td align="center">β</td>
|
| 227 |
+
<td align="center">β</td>
|
| 228 |
+
<td align="center">β</td>
|
| 229 |
<td align="center">70.2</td>
|
| 230 |
+
<td align="center">74.8</td>
|
| 231 |
+
<td align="center">β</td>
|
| 232 |
+
<td align="center">β</td>
|
| 233 |
+
<td align="center">β</td>
|
| 234 |
</tr>
|
| 235 |
<tr>
|
| 236 |
+
<td>InternVL2.5β78B</td>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 237 |
<td align="center">β
</td>
|
| 238 |
+
<td align="center">β</td>
|
| 239 |
+
<td align="center">β</td>
|
| 240 |
+
<td align="center">β</td>
|
| 241 |
+
<td align="center">β</td>
|
| 242 |
+
<td align="center">β</td>
|
| 243 |
+
<td align="center">β</td>
|
| 244 |
+
<td align="center">β</td>
|
| 245 |
+
<td align="center">β</td>
|
| 246 |
<td align="center">70.1</td>
|
| 247 |
+
<td align="center">72.3</td>
|
| 248 |
+
<td align="center">β</td>
|
| 249 |
+
<td align="center">33.2</td>
|
| 250 |
+
<td align="center">β</td>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 251 |
</tr>
|
| 252 |
</tbody>
|
| 253 |
</table>
|
| 254 |
|
| 255 |
|
| 256 |
+
|
| 257 |
---
|
| 258 |
|
| 259 |
|