Replace model card with Cortex-Mini-1-Preview README
Browse files
README.md
CHANGED
|
@@ -1,582 +1,163 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
|
| 108 |
-
|
| 109 |
-
|
| 110 |
-
|
| 111 |
-
|
| 112 |
-
|
| 113 |
-
|
| 114 |
-
|
| 115 |
-
|
| 116 |
-
|
| 117 |
-
|
| 118 |
-
|
| 119 |
-
|
| 120 |
-
|
| 121 |
-
|
| 122 |
-
|
| 123 |
-
|
| 124 |
-
|
| 125 |
-
|
| 126 |
-
|
| 127 |
-
|
| 128 |
-
|
| 129 |
-
|
| 130 |
-
|
| 131 |
-
|
| 132 |
-
|
| 133 |
-
|
| 134 |
-
|
| 135 |
-
|
| 136 |
-
|
| 137 |
-
|
| 138 |
-
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
|
| 142 |
-
|
| 143 |
-
|
| 144 |
-
|
| 145 |
-
|
| 146 |
-
|
| 147 |
-
|
| 148 |
-
|
| 149 |
-
|
| 150 |
-
|
| 151 |
-
|
| 152 |
-
|
| 153 |
-
|
| 154 |
-
|
| 155 |
-
|
| 156 |
-
|
| 157 |
-
|
| 158 |
-
|
| 159 |
-
|
| 160 |
-
|
| 161 |
-
|
| 162 |
-
|
| 163 |
-
|
| 164 |
-
</tr>
|
| 165 |
-
<tr>
|
| 166 |
-
<td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Competitive coding</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">LiveCodeBench v6</div></td>
|
| 167 |
-
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>90.3</strong></td>
|
| 168 |
-
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">83.9</td>
|
| 169 |
-
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">89.6</td>
|
| 170 |
-
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td>
|
| 171 |
-
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">88.8</td>
|
| 172 |
-
</tr>
|
| 173 |
-
</tbody>
|
| 174 |
-
</table>
|
| 175 |
-
<div style="margin-top:12px;font-size:11px;line-height:1.5;color:rgba(0,0,0,0.72)">
|
| 176 |
-
<ol style="margin:0;padding-left:20px">
|
| 177 |
-
<li>SWE-bench Pro: Except for Opus4.6 Max, which uses the officially reported score, all models are evaluated with the Claude Code harness at temp=1.0, top_p=0.95, and a 256K context window. Problematic tasks were corrected, and all baseline models were re-evaluated on the refined benchmark.</li>
|
| 178 |
-
<li>NL2Repo-Bench: Evaluated with the Claude Code harness. To prevent reward hacking, we disable Bash commands that attempt to access the specific repository, such as pip download, pip install, and git clone.</li>
|
| 179 |
-
<li>DeepSWE 1.1: Evaluated with the Claude Code harness at temp=1.0, top_p=0.95, and a 256K context window.</li>
|
| 180 |
-
<li>QwenSWEBench: In-house coding benchmark for evaluating models' software engineering capabilities. Evaluated with the Claude Code harness. Reporting avg@3 with an 8-hour timeout, max_tokens=32,768, temperature=1.0, and a 256K context window.</li>
|
| 181 |
-
<li>CoWorkBench: In-house cowork benchmark for evaluating long-horizon tasks across computer science, finance, law, medical, and other productivity domains.</li>
|
| 182 |
-
<li>HLE: Judged by GPT-4o.</li>
|
| 183 |
-
<li>The best result in each row is shown in bold.</li>
|
| 184 |
-
<li>Empty cells (--) indicate that results are not yet available or not applicable.</li>
|
| 185 |
-
</ol>
|
| 186 |
-
</div>
|
| 187 |
-
</div>
|
| 188 |
-
|
| 189 |
-
### VL Performance
|
| 190 |
-
<div style="font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif;max-width:1200px;margin:0 auto;padding:16px 0">
|
| 191 |
-
<table class="vl-table" style="width:100%;table-layout:fixed;border-collapse:collapse;font-size:13px">
|
| 192 |
-
<thead><tr><th style="padding:10px 7px;text-align:left;font-weight:600;border-bottom:2px solid #0A2EFE;color:#0A2EFE"></th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #0A2EFE;color:#0A2EFE;font-size: 14px;width:14.00%;background:rgba(10, 46, 254, 0.08);">Qwen3.8-27B</th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #0A2EFE;color:#0A2EFE;font-size: 14px;width:14.00%;">Qwen3.6-27B</th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #0A2EFE;color:#0A2EFE;font-size: 14px;width:14.00%;">Qwen3.7-Plus</th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #0A2EFE;color:#0A2EFE;font-size: 14px;width:14.00%;">Muse Glimmer-30B</th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #0A2EFE;color:#0A2EFE;font-size: 14px;width:14.00%;">Opus4.6 Max</th></tr></thead>
|
| 193 |
-
<tbody>
|
| 194 |
-
<tr><td colspan="6" style="padding:8px 12px;font-weight:600;color:#0A2EFE;border-bottom:1px solid rgba(10, 46, 254, 0.2);background:#D6DAFC">Agentic Multimodal Intelligence</td></tr>
|
| 195 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Computer use</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">OSWorld-Verified</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>84.3</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">63.9</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">73.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">65.9</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">72.7</td></tr>
|
| 196 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Browser use</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">WebArena-Verified</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>64.8</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">48.8</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">55.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td></tr>
|
| 197 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Mobile use</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">AndroidWorld</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>81.9</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">70.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">81.0</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">62.0</td></tr>
|
| 198 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Application recreation</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">RecreationBench</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>47.1</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">29.8</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">30.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td></tr>
|
| 199 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Multimodal tool use</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">ClawEval-MM</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Pass@3</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>57.4</strong></div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Average</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">56.9</div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Pass@3</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">42.6</div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Average</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">50.4</div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Pass@3</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>57.4</strong></div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Average</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>60.1</strong></div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Pass@3</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">52.5</div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Average</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">54.7</div></div></div></td></tr>
|
| 200 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Multimodal software engineering</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">SWE-MM</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>38.6</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">25.7</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">30.0</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">27.1</td></tr>
|
| 201 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Visual web development</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">Vision2Web</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>62.9</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">45.0</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">42.1</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td></tr>
|
| 202 |
-
<tr><td colspan="6" style="padding:8px 12px;font-weight:600;color:#0A2EFE;border-bottom:1px solid rgba(10, 46, 254, 0.2);background:#D6DAFC">General Multimodal Intelligence</td></tr>
|
| 203 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Visual math problem solving</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">MathVision</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">90.0</div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">With CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>94.6</strong></div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">85.1</div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>90.3</strong></div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">65.5</div></div></div></td></tr>
|
| 204 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">General visual reasoning</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">BabyVision</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>65.7</strong></div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">With CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>85.6</strong></div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">28.9</div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">64.7</div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">With CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">70.4</div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">12.6</div></div></div></td></tr>
|
| 205 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Scientific chart analysis</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">CharXiv (RQ)</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">83.7</div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">With CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>90.2</strong></div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">78.4</div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717"><strong>85.8</strong></div></div><div style="margin-top:7px"><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">With CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">85.9</div></div></div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">78.8</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><div class="metric-stack" style="padding:3px 0"><div><div class="metric-label" style="font-size:10px;font-weight:400;line-height:1.1;color:#777">Without CI</div><div class="metric-value" style="margin-top:2px;font-size:15px;line-height:1.15;color:#171717">66.0</div></div></div></td></tr>
|
| 206 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Document intelligence</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">OmniDocBench 1.5</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;">91.1</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">89.4</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>91.4</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">75.8</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">86.6</td></tr>
|
| 207 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Real-world perception</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">RealWorldQA</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;">85.9</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">84.1</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>86.9</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">73.9</td></tr>
|
| 208 |
-
<tr><td class="benchmark-cell" style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:15px;font-weight:600;line-height:1.22;color:#171717">Embodied intelligence</div><div class="benchmark-name" style="margin-top:4px;font-size:11px;font-weight:400;line-height:1.2;color:#6B6B6B">ERQA</div></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(10, 46, 254, 0.08);vertical-align:middle;font-size:15px;line-height:1.2;">65.5</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">62.5</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;"><strong>69.8</strong></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">--</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:15px;line-height:1.2;">40.8</td></tr>
|
| 209 |
-
</tbody>
|
| 210 |
-
</table>
|
| 211 |
-
<div style="margin-top:12px;font-size:11px;line-height:1.5;color:rgba(0,0,0,0.72)">
|
| 212 |
-
<ol style="margin:0;padding-left:20px">
|
| 213 |
-
<li>MathVision, BabyVision, and CharXiv (RQ): Where both settings are available, cells report “Without CI” and “With CI” separately; otherwise, only the available setting is shown. A small number of incorrect ground-truth annotations in MathVision and CharXiv (RQ) were corrected following manual verification, and all reported scores on those benchmarks were computed using the corrected annotations.</li>
|
| 214 |
-
<li>MathVision: Qwen3.8-27B is evaluated using the fixed prompt: “Please reason step by step, and put your final answer within <code>\boxed{}</code>.” For the remaining models, we report the higher score from two prompt variants—one with and one without the <code>\boxed{}</code> formatting requirement.</li>
|
| 215 |
-
<li>WebArena-Verified: Scores are computed with the official WebArena-Verified grader under the OSWorld scaffold.</li>
|
| 216 |
-
<li>RecreationBench: An in-house, long-horizon application-recreation benchmark designed to evaluate hybrid-agent capabilities across five platforms: desktop (Ubuntu, macOS, and Windows), mobile (Android), and the web.</li>
|
| 217 |
-
<li>ClawEval-MM: Scores are reported as “Pass@3 / average score.” Pass@3 is the percentage of tasks passed in at least one of three trials; the average score is the mean benchmark score across the three trials.</li>
|
| 218 |
-
<li>Vision2Web: Scores are averaged across the frontend, webpage, and website categories. Evaluations use the Claude Code harness and are judged by <code>gpt-5.4-2026-03-05</code>.</li>
|
| 219 |
-
<li>SWE-MM: Scores are evaluated on the Claude Code harness using the public dev split of SWE-bench Multimodal, with the modifications described in Appendix 8.3 of the Claude Opus 4.7 system card.</li>
|
| 220 |
-
<li>Empty cells (--) indicate that results are not yet available or not applicable.</li>
|
| 221 |
-
</ol></div>
|
| 222 |
-
</div>
|
| 223 |
-
|
| 224 |
-
|
| 225 |
-
## Quickstart
|
| 226 |
-
|
| 227 |
-
For streamlined integration, we recommend using Qwen3.8 via APIs.
|
| 228 |
-
|
| 229 |
-
### Serving Qwen3.8
|
| 230 |
-
|
| 231 |
-
> [!Important]
|
| 232 |
-
> Inference efficiency and throughput vary significantly across frameworks.
|
| 233 |
-
> We recommend using the latest framework versions to ensure optimal performance and compatibility.
|
| 234 |
-
> For production workloads or high-throughput scenarios, dedicated serving engines such as SGLang, vLLM, or TokenSpeed are recommended.
|
| 235 |
-
|
| 236 |
-
Qwen3.8 can be deployed with popular inference frameworks, e.g.:
|
| 237 |
-
|
| 238 |
-
- [SGLang](https://www.sglang.io/): [Qwen3.8 Cookbook](https://docs.sglang.io/cookbook/autoregressive/Qwen/Qwen3.8-27B)
|
| 239 |
-
- [vLLM](https://vllm.ai/): [Qwen3.8 Recipe](https://recipes.vllm.ai/Qwen/Qwen3.8-27B)
|
| 240 |
-
- [TokenSpeed](https://lightseek.org/tokenspeed/): [Qwen3.8 Recipe](https://lightseek.org/tokenspeed/recipes/models#qwen3-8)
|
| 241 |
-
|
| 242 |
-
|
| 243 |
-
### API Usage
|
| 244 |
-
|
| 245 |
-
> [!Important]
|
| 246 |
-
> Qwen3.8 models operate in thinking mode by default, generating thinking content signified by `<think>\n...</think>\n\n` before producing the final response.
|
| 247 |
-
> To disable thinking content and obtain a direct response, refer to the examples [here](#instruct-or-non-thinking-mode).
|
| 248 |
-
|
| 249 |
-
|
| 250 |
-
> [!Tip]
|
| 251 |
-
> We recommend using the following sets of sampling parameters for generation:
|
| 252 |
-
> - Thinking Mode: `temperature=1.0`, `top_p=0.95`, `top_k=20`, `min_p=0.0`, `presence_penalty=0.0`, `repetition_penalty=1.0`
|
| 253 |
-
> - Instruct (or non-thinking) mode: `temperature=0.7`, `top_p=0.80`, `top_k=20`, `min_p=0.0`, `presence_penalty=1.5`, `repetition_penalty=1.0`
|
| 254 |
-
>
|
| 255 |
-
> Please note that the support for sampling parameters varies according to inference frameworks.
|
| 256 |
-
|
| 257 |
-
|
| 258 |
-
Qwen3.8 comes with official support for `reasoning_effort`, which can be used to adjust reasoning depth and control cost:
|
| 259 |
-
- `xhigh` (default): for complex tasks demanding thorough analysis
|
| 260 |
-
- `medium`: balancing accuracy and speed
|
| 261 |
-
- `low`: efficient reasoning optimizing for speed and cost
|
| 262 |
-
|
| 263 |
-
|
| 264 |
-
In addition, `preserve_thinking` is enabled by default for all workloads for the best out-of-the-box experience. To disable preserved thinking, refer to the examples [here](#disable-preserved-thinking).
|
| 265 |
-
|
| 266 |
-
> [!Tip]
|
| 267 |
-
> In multi-turn agentic tasks, lower reasoning effort does not always reduce overall task completion time. Although it may produce faster per-turn responses, it can also lead to insufficient analysis, more failures, and repeated retries, which may increase total latency and token consumption.
|
| 268 |
-
|
| 269 |
-
|
| 270 |
-
#### Chat Completions API
|
| 271 |
-
|
| 272 |
-
The Chat Completions API can be used with most inference frameworks, as well as [Qwen Cloud](https://www.qwencloud.com/).
|
| 273 |
-
Before starting, make sure the OpenAI Python SDK is installed and the API key and the API base URL are configured, e.g.:
|
| 274 |
-
```shell
|
| 275 |
-
pip install -U openai
|
| 276 |
-
|
| 277 |
-
# Set the following accordingly
|
| 278 |
-
export OPENAI_BASE_URL='your-base-url'
|
| 279 |
-
export OPENAI_API_KEY='your-api-key'
|
| 280 |
-
```
|
| 281 |
-
|
| 282 |
-
##### Text-Only Input
|
| 283 |
-
|
| 284 |
-
```python
|
| 285 |
-
from openai import OpenAI
|
| 286 |
-
# Configured by environment variables
|
| 287 |
-
client = OpenAI()
|
| 288 |
-
|
| 289 |
-
messages = [{"role": "user", "content": "Write a Python function to merge two sorted linked lists."}]
|
| 290 |
-
|
| 291 |
-
completion = client.chat.completions.create(
|
| 292 |
-
model="Qwen/Qwen3.8-27B",
|
| 293 |
-
messages=messages,
|
| 294 |
-
extra_body={
|
| 295 |
-
"chat_template_kwargs": {
|
| 296 |
-
"enable_thinking": True, # on by default
|
| 297 |
-
"preserve_thinking": True, # on by default
|
| 298 |
-
},
|
| 299 |
-
},
|
| 300 |
-
reasoning_effort="xhigh", # xhigh by default; supported levels are xhigh, medium, and low
|
| 301 |
-
stream=True,
|
| 302 |
-
stream_options={"include_usage": True},
|
| 303 |
-
)
|
| 304 |
-
|
| 305 |
-
reasoning_content = ""
|
| 306 |
-
answer_content = ""
|
| 307 |
-
is_answering = False
|
| 308 |
-
print("\n" + "=" * 20 + "Reasoning" + "=" * 20 + "\n")
|
| 309 |
-
|
| 310 |
-
for chunk in completion:
|
| 311 |
-
if not chunk.choices:
|
| 312 |
-
print("\nUsage:")
|
| 313 |
-
print(chunk.usage)
|
| 314 |
-
continue
|
| 315 |
-
|
| 316 |
-
delta = chunk.choices[0].delta
|
| 317 |
-
|
| 318 |
-
if hasattr(delta, "reasoning_content") and delta.reasoning_content is not None:
|
| 319 |
-
if not is_answering:
|
| 320 |
-
print(delta.reasoning_content, end="", flush=True)
|
| 321 |
-
reasoning_content += delta.reasoning_content
|
| 322 |
-
elif hasattr(delta, "reasoning") and delta.reasoning is not None:
|
| 323 |
-
if not is_answering:
|
| 324 |
-
print(delta.reasoning, end="", flush=True)
|
| 325 |
-
reasoning_content += delta.reasoning
|
| 326 |
-
|
| 327 |
-
if hasattr(delta, "content") and delta.content:
|
| 328 |
-
if not is_answering:
|
| 329 |
-
print("\n" + "=" * 20 + "Answer" + "=" * 20 + "\n")
|
| 330 |
-
is_answering = True
|
| 331 |
-
print(delta.content, end="", flush=True)
|
| 332 |
-
answer_content += delta.content
|
| 333 |
-
|
| 334 |
-
messages.append({
|
| 335 |
-
"role": "assistant",
|
| 336 |
-
"content": answer_content,
|
| 337 |
-
"reasoning_content": reasoning_content,
|
| 338 |
-
"reasoning": reasoning_content,
|
| 339 |
-
})
|
| 340 |
-
```
|
| 341 |
-
|
| 342 |
-
|
| 343 |
-
##### Image Input
|
| 344 |
-
|
| 345 |
-
```python
|
| 346 |
-
from openai import OpenAI
|
| 347 |
-
# Configured by environment variables
|
| 348 |
-
client = OpenAI()
|
| 349 |
-
|
| 350 |
-
messages = [
|
| 351 |
-
{
|
| 352 |
-
"role": "user",
|
| 353 |
-
"content": [
|
| 354 |
-
{
|
| 355 |
-
"type": "image_url",
|
| 356 |
-
"image_url": {
|
| 357 |
-
"url": "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3.5/demo/CI_Demo/mathv-1327.jpg"
|
| 358 |
-
}
|
| 359 |
-
},
|
| 360 |
-
{
|
| 361 |
-
"type": "text",
|
| 362 |
-
"text": "The centres of the four illustrated circles are in the corners of the square. The two big circles touch each other and also the two little circles. With which factor do you have to multiply the radii of the little circles to obtain the radius of the big circles?\nChoices:\n(A) $\\frac{2}{9}$\n(B) $\\sqrt{5}$\n(C) $0.8 \\cdot \\pi$\n(D) 2.5\n(E) $1+\\sqrt{2}$"
|
| 363 |
-
}
|
| 364 |
-
]
|
| 365 |
-
}
|
| 366 |
-
]
|
| 367 |
-
|
| 368 |
-
chat_response = client.chat.completions.create(
|
| 369 |
-
model="Qwen/Qwen3.8-27B",
|
| 370 |
-
messages=messages,
|
| 371 |
-
)
|
| 372 |
-
print("Chat response:", chat_response)
|
| 373 |
-
```
|
| 374 |
-
|
| 375 |
-
##### Video Input
|
| 376 |
-
|
| 377 |
-
```python
|
| 378 |
-
from openai import OpenAI
|
| 379 |
-
# Configured by environment variables
|
| 380 |
-
client = OpenAI()
|
| 381 |
-
|
| 382 |
-
messages = [
|
| 383 |
-
{
|
| 384 |
-
"role": "user",
|
| 385 |
-
"content": [
|
| 386 |
-
{
|
| 387 |
-
"type": "video_url",
|
| 388 |
-
"video_url": {
|
| 389 |
-
"url": "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3.5/demo/video/N1cdUjctpG8.mp4"
|
| 390 |
-
}
|
| 391 |
-
},
|
| 392 |
-
{
|
| 393 |
-
"type": "text",
|
| 394 |
-
"text": "How many porcelain jars were discovered in the niches located in the primary chamber of the tomb?"
|
| 395 |
-
}
|
| 396 |
-
]
|
| 397 |
-
}
|
| 398 |
-
]
|
| 399 |
-
|
| 400 |
-
chat_response = client.chat.completions.create(
|
| 401 |
-
model="Qwen/Qwen3.8-27B",
|
| 402 |
-
messages=messages,
|
| 403 |
-
)
|
| 404 |
-
|
| 405 |
-
# When vLLM is launched with `--media-io-kwargs '{"video": {"num_frames": -1}}'`,
|
| 406 |
-
# video frame sampling can be configured via `extra_body` (e.g., by setting `fps`).
|
| 407 |
-
# This feature is currently supported only in vLLM.
|
| 408 |
-
#
|
| 409 |
-
# By default, `fps=2` and `do_sample_frames=True`.
|
| 410 |
-
# With `do_sample_frames=True`, you can customize the `fps` value to set your desired video sampling rate.
|
| 411 |
-
# chat_response = client.chat.completions.create(
|
| 412 |
-
# model="Qwen/Qwen3.8-27B",
|
| 413 |
-
# messages=messages,
|
| 414 |
-
# extra_body={
|
| 415 |
-
# "mm_processor_kwargs": {"fps": 2, "do_sample_frames": True},
|
| 416 |
-
# },
|
| 417 |
-
# )
|
| 418 |
-
|
| 419 |
-
print("Chat response:", chat_response)
|
| 420 |
-
```
|
| 421 |
-
|
| 422 |
-
|
| 423 |
-
##### Instruct (or Non-Thinking) Mode
|
| 424 |
-
|
| 425 |
-
Qwen3.8-27B will think by default before responding.
|
| 426 |
-
You can obtain a direct response from the model without thinking by configuring the API parameters.
|
| 427 |
-
For example,
|
| 428 |
-
```python
|
| 429 |
-
from openai import OpenAI
|
| 430 |
-
# Configured by environment variables
|
| 431 |
-
client = OpenAI()
|
| 432 |
-
|
| 433 |
-
messages = [
|
| 434 |
-
{
|
| 435 |
-
"role": "user",
|
| 436 |
-
"content": [
|
| 437 |
-
{
|
| 438 |
-
"type": "image_url",
|
| 439 |
-
"image_url": {
|
| 440 |
-
"url": "https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3.5/demo/RealWorld/RealWorld-04.png"
|
| 441 |
-
}
|
| 442 |
-
},
|
| 443 |
-
{
|
| 444 |
-
"type": "text",
|
| 445 |
-
"text": "Where is this?"
|
| 446 |
-
}
|
| 447 |
-
]
|
| 448 |
-
}
|
| 449 |
-
]
|
| 450 |
-
|
| 451 |
-
chat_response = client.chat.completions.create(
|
| 452 |
-
model="Qwen/Qwen3.8-27B",
|
| 453 |
-
messages=messages,
|
| 454 |
-
temperature=0.7,
|
| 455 |
-
top_p=0.8,
|
| 456 |
-
presence_penalty=1.5,
|
| 457 |
-
extra_body={
|
| 458 |
-
"top_k": 20,
|
| 459 |
-
"chat_template_kwargs": {"enable_thinking": False},
|
| 460 |
-
},
|
| 461 |
-
)
|
| 462 |
-
print("Chat response:", chat_response)
|
| 463 |
-
```
|
| 464 |
-
|
| 465 |
-
> [!Note]
|
| 466 |
-
> If you are using APIs from Qwen Cloud, in addition to changing `model`, please use `"enable_thinking": False` instead of `"chat_template_kwargs": {"enable_thinking": False}`.
|
| 467 |
-
|
| 468 |
-
|
| 469 |
-
##### Disable Preserved Thinking
|
| 470 |
-
|
| 471 |
-
|
| 472 |
-
By default, Qwen3.8 retains thinking blocks from all historical messages, maintaining a complete reasoning trace across the conversation. This behavior, known as preserved thinking, ensures full context continuity and is especially beneficial for agent scenarios where decision consistency and reduced redundant reasoning are critical. It also improves KV cache utilization, optimizing inference efficiency in both thinking and non-thinking modes.
|
| 473 |
-
|
| 474 |
-
If you prefer to retain only the thinking blocks from the latest user message, you can disable this behavior by setting `preserve_thinking` to `False`:
|
| 475 |
-
|
| 476 |
-
```python
|
| 477 |
-
from openai import OpenAI
|
| 478 |
-
|
| 479 |
-
# Configured by environment variables
|
| 480 |
-
client = OpenAI()
|
| 481 |
-
messages = [...]
|
| 482 |
-
chat_response = client.chat.completions.create(
|
| 483 |
-
model="Qwen/Qwen3.8-27B",
|
| 484 |
-
messages=messages,
|
| 485 |
-
extra_body={
|
| 486 |
-
"chat_template_kwargs": {"preserve_thinking": False},
|
| 487 |
-
},
|
| 488 |
-
)
|
| 489 |
-
print("Chat response:", chat_response)
|
| 490 |
-
```
|
| 491 |
-
|
| 492 |
-
> [!Note]
|
| 493 |
-
> If you are using APIs from Qwen Cloud, in addition to changing `model`, please use `"preserve_thinking": False` directly instead of wrapping it in `chat_template_kwargs`.
|
| 494 |
-
|
| 495 |
-
|
| 496 |
-
## Best Practices
|
| 497 |
-
|
| 498 |
-
To achieve optimal performance, we recommend the following settings:
|
| 499 |
-
|
| 500 |
-
1. **Sampling Parameters**: We suggest using the following sets of sampling parameters:
|
| 501 |
-
|
| 502 |
-
- Thinking Mode: `temperature=1.0`, `top_p=0.95`, `top_k=20`, `min_p=0.0`, `presence_penalty=0.0`, `repetition_penalty=1.0`
|
| 503 |
-
- Instruct (or non-thinking) mode: `temperature=0.7`, `top_p=0.80`, `top_k=20`, `min_p=0.0`, `presence_penalty=1.5`, `repetition_penalty=1.0`
|
| 504 |
-
|
| 505 |
-
For supported frameworks, you can adjust the `presence_penalty` parameter between 0 and 2 to reduce endless repetition. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.
|
| 506 |
-
|
| 507 |
-
2. **Adequate Output Length**: To optimize performance on agentic tasks, we recommend allocating sufficient output length to allow the model to generate detailed and comprehensive responses. For frameworks that support separate token limits for internal reasoning and final outputs, we suggest the following configuration within the 1M context length:
|
| 508 |
-
|
| 509 |
-
- Reasoning Content: Set the maximum output length to 262,144 tokens.
|
| 510 |
-
- Final Response: Set the maximum output length to 131,072 tokens.
|
| 511 |
-
|
| 512 |
-
These settings provide the necessary capacity for complex reasoning while ensuring ample space for high-quality final deliverables.
|
| 513 |
-
|
| 514 |
-
3. **Processing Ultra-Long Texts**: Qwen3.8-27B natively supports context lengths of up to 262,144 tokens. For long-horizon tasks where the total length (including both input and output) exceeds this limit, we recommend using RoPE scaling techniques to handle long texts effectively, e.g., YaRN.
|
| 515 |
-
|
| 516 |
-
YaRN is currently supported by several inference frameworks, e.g., vLLM, SGLang, and TokenSpeed.
|
| 517 |
-
In general, there are two approaches to enabling YaRN for supported frameworks:
|
| 518 |
-
|
| 519 |
-
- Modifying the model configuration file:
|
| 520 |
-
|
| 521 |
-
In the `config.json` file, change the `rope_parameters` fields in `text_config` to:
|
| 522 |
-
```json
|
| 523 |
-
{
|
| 524 |
-
"mrope_interleaved": true,
|
| 525 |
-
"mrope_section": [
|
| 526 |
-
11,
|
| 527 |
-
11,
|
| 528 |
-
10
|
| 529 |
-
],
|
| 530 |
-
"rope_type": "yarn",
|
| 531 |
-
"rope_theta": 10000000,
|
| 532 |
-
"partial_rotary_factor": 0.25,
|
| 533 |
-
"factor": 4.0,
|
| 534 |
-
"original_max_position_embeddings": 262144,
|
| 535 |
-
}
|
| 536 |
-
```
|
| 537 |
-
|
| 538 |
-
- Passing command line arguments:
|
| 539 |
-
|
| 540 |
-
For vLLM, you can use
|
| 541 |
-
```shell
|
| 542 |
-
VLLM_ALLOW_LONG_MAX_MODEL_LEN=1 vllm serve ... --hf-overrides '{"text_config": {"rope_parameters": {"mrope_interleaved": true, "mrope_section": [11, 11, 10], "rope_type": "yarn", "rope_theta": 10000000, "partial_rotary_factor": 0.25, "factor": 4.0, "original_max_position_embeddings": 262144}}}' --max-model-len 1000000
|
| 543 |
-
```
|
| 544 |
-
|
| 545 |
-
For SGLang, you can use
|
| 546 |
-
```shell
|
| 547 |
-
SGLANG_ALLOW_OVERWRITE_LONGER_CONTEXT_LEN=1 python -m sglang.launch_server ... --json-model-override-args '{"text_config": {"rope_parameters": {"mrope_interleaved": true, "mrope_section": [11, 11, 10], "rope_type": "yarn", "rope_theta": 10000000, "partial_rotary_factor": 0.25, "factor": 4.0, "original_max_position_embeddings": 262144}}}' --context-length 1000000
|
| 548 |
-
```
|
| 549 |
-
|
| 550 |
-
For TokenSpeed, you can use
|
| 551 |
-
```shell
|
| 552 |
-
TOKENSPEED_ALLOW_OVERWRITE_LONGER_CONTEXT_LEN=1 tokenspeed serve ... --hf-overrides '{"text_config": {"rope_parameters": {"mrope_interleaved": true, "mrope_section": [11, 11, 10], "rope_type": "yarn", "rope_theta": 10000000, "partial_rotary_factor": 0.25, "factor": 4.0, "original_max_position_embeddings": 262144}}}' --max-model-len 1000000
|
| 553 |
-
```
|
| 554 |
-
|
| 555 |
-
> [!NOTE]
|
| 556 |
-
> All the notable open-source frameworks implement static YaRN, which means the scaling factor remains constant regardless of input length, **potentially impacting performance on shorter texts.**
|
| 557 |
-
> We advise modifying the `rope_parameters` configuration only when processing long contexts is required.
|
| 558 |
-
> It is also recommended to modify the `factor` as needed. For example, if the typical context length for your application is 524,288 tokens, it would be better to set `factor` as 2.0.
|
| 559 |
-
|
| 560 |
-
|
| 561 |
-
4. **Long Video Understanding**: To optimize inference efficiency for plain text and images, the `size` parameter in the released `video_preprocessor_config.json` is conservatively configured. It is recommended to set the `longest_edge` parameter in the video_preprocessor_config file to 469,762,048 (corresponding to 224k video tokens) to enable higher frame-rate sampling for hour-scale videos and thereby achieve superior performance. For example,
|
| 562 |
-
```json
|
| 563 |
-
{"longest_edge": 469762048, "shortest_edge": 4096}
|
| 564 |
-
```
|
| 565 |
-
|
| 566 |
-
Alternatively, override the default values via engine startup parameters. For implementation details, refer to: [vLLM](https://github.com/vllm-project/vllm/pull/34330) / [SGLang](https://github.com/sgl-project/sglang/pull/18467).
|
| 567 |
-
|
| 568 |
-
|
| 569 |
-
## Citation
|
| 570 |
-
|
| 571 |
-
If you find our work helpful, feel free to give us a cite.
|
| 572 |
-
|
| 573 |
-
|
| 574 |
-
```bibtex
|
| 575 |
-
@misc{qwen38,
|
| 576 |
-
title = {{Qwen3.8-Max}: A New Bar for Coding and Cowork},
|
| 577 |
-
url = {https://qwen.ai/blog?id=qwen3.8},
|
| 578 |
-
author = {{Qwen Team}},
|
| 579 |
-
month = {August},
|
| 580 |
-
year = {2026}
|
| 581 |
-
}
|
| 582 |
-
```
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
license_name: apache-2.0
|
| 4 |
+
license_link: https://www.apache.org/licenses/LICENSE-2.0
|
| 5 |
+
library_name: transformers
|
| 6 |
+
pipeline_tag: image-text-to-text
|
| 7 |
+
language:
|
| 8 |
+
- en
|
| 9 |
+
- zh
|
| 10 |
+
base_model: Qwen/Qwen3.8-27B
|
| 11 |
+
base_model_relation: finetune
|
| 12 |
+
tags:
|
| 13 |
+
- qwen
|
| 14 |
+
- qwen3.8
|
| 15 |
+
- cortex
|
| 16 |
+
- cortexlm
|
| 17 |
+
- post-training
|
| 18 |
+
- vision-language
|
| 19 |
+
- conversational
|
| 20 |
+
- preview
|
| 21 |
+
---
|
| 22 |
+
|
| 23 |
+
<p align="center">
|
| 24 |
+
<img src="banner.jpg" alt="Cortex" width="100%" />
|
| 25 |
+
</p>
|
| 26 |
+
|
| 27 |
+
# Cortex-Mini-1-Preview
|
| 28 |
+
|
| 29 |
+
**Cortex-Mini-1-Preview** is a post-trained derivative of [Qwen/Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B), released by [CortexLM](https://huggingface.co/CortexLM).
|
| 30 |
+
|
| 31 |
+
This repository is a fork of the official Qwen3.8-27B checkpoint. We start from that open-weight native vision-language model and continue with Cortex post-training (Relearn). The goal is to improve the model on held-out tasks **without** collapsing general capability or overfitting public eval splits.
|
| 32 |
+
|
| 33 |
+
> Preview. Weights, training recipe, and numbers in this card will move. Treat this page as the public home for the Cortex Mini 1 line, not a frozen release.
|
| 34 |
+
|
| 35 |
+
## Thanks
|
| 36 |
+
|
| 37 |
+
This work exists because the Qwen team open-sourced [Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B) under Apache 2.0.
|
| 38 |
+
|
| 39 |
+
Thank you to the **Qwen Team** and **Alibaba Cloud** for the architecture, the training, the weights, and the docs. Cortex-Mini-1-Preview inherits their model; our contribution is post-training on top. If you use this checkpoint, please also cite Qwen3.8 (see [Citation](#citation)).
|
| 40 |
+
|
| 41 |
+
## What this is
|
| 42 |
+
|
| 43 |
+
| | |
|
| 44 |
+
|---|---|
|
| 45 |
+
| Base | [`Qwen/Qwen3.8-27B`](https://huggingface.co/Qwen/Qwen3.8-27B) |
|
| 46 |
+
| Relation | Post-training / continued training (`finetune`) |
|
| 47 |
+
| Type | Native vision-language causal LM (text, image, video) |
|
| 48 |
+
| Parameters | 27B dense |
|
| 49 |
+
| License | Apache License 2.0 (same as the base) |
|
| 50 |
+
| Org | [CortexLM](https://huggingface.co/CortexLM) |
|
| 51 |
+
|
| 52 |
+
Qwen3.8-27B already accepts image and video input. Cortex-Mini-1-Preview does **not** bolt on a separate vision encoder. Miners and researchers post-train this 27B checkpoint itself.
|
| 53 |
+
|
| 54 |
+
## License
|
| 55 |
+
|
| 56 |
+
The base model is licensed under **Apache License 2.0**, Copyright 2026 Alibaba Cloud. That license allows commercial use, modification, and redistribution, including this fork.
|
| 57 |
+
|
| 58 |
+
This repository redistributes Qwen3.8-27B weights and configuration as a derivative work:
|
| 59 |
+
|
| 60 |
+
- The Apache 2.0 license text ships with the repo (`LICENSE`).
|
| 61 |
+
- Attribution to the Qwen Team / Alibaba Cloud is retained (this card + `NOTICE`).
|
| 62 |
+
- Files we change are marked as such. The model card, banner, and Cortex post-training artifacts are new.
|
| 63 |
+
|
| 64 |
+
You must keep the license, attribution, and NOTICE if you redistribute further. "Qwen" remains a trademark of the original authors; this project does not claim that mark.
|
| 65 |
+
|
| 66 |
+
Full terms: [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0) and the `LICENSE` file in this repo.
|
| 67 |
+
|
| 68 |
+
## Model overview (inherited from Qwen3.8-27B)
|
| 69 |
+
|
| 70 |
+
These architectural facts come from the [Qwen3.8-27B model card](https://huggingface.co/Qwen/Qwen3.8-27B). Cortex post-training does not change the network shape.
|
| 71 |
+
|
| 72 |
+
- **Type:** Causal language model with vision encoder (native VLM)
|
| 73 |
+
- **Training stage (base):** Pre-training and post-training by Qwen
|
| 74 |
+
- **Training stage (this repo):** Additional Cortex post-training on the Qwen checkpoint
|
| 75 |
+
- **Parameters:** 27B
|
| 76 |
+
- **Hidden size:** 5120
|
| 77 |
+
- **Layers:** 64
|
| 78 |
+
- **Context:** 262,144 tokens natively, extensible to 1,000,000 with YaRN
|
| 79 |
+
- **Modalities:** Text, image, video
|
| 80 |
+
|
| 81 |
+
For serving details, sampling defaults, thinking mode, and YaRN flags, follow the Qwen card. Those recipes still apply.
|
| 82 |
+
|
| 83 |
+
## Intended use
|
| 84 |
+
|
| 85 |
+
- Research and development on Cortex Relearn (post-train a capable open VLM without wrecking general skills).
|
| 86 |
+
- Downstream chat, coding, agent, and vision-language workloads that already run on Qwen3.8-27B.
|
| 87 |
+
- Further fine-tuning under Apache 2.0.
|
| 88 |
+
|
| 89 |
+
**Out of scope**
|
| 90 |
+
|
| 91 |
+
- Do not present this preview as a drop-in Qwen Cloud / Qwen3.8-Max replacement.
|
| 92 |
+
- Do not treat unpublished scores as Qwen's official numbers.
|
| 93 |
+
- Do not strip license or attribution.
|
| 94 |
+
|
| 95 |
+
## How to use
|
| 96 |
+
|
| 97 |
+
Load the repo id once weights are in place:
|
| 98 |
+
|
| 99 |
+
```python
|
| 100 |
+
from transformers import AutoModelForImageTextToText, AutoProcessor
|
| 101 |
+
|
| 102 |
+
model_id = "CortexLM/Cortex-Mini-1-Preview"
|
| 103 |
+
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
|
| 104 |
+
model = AutoModelForImageTextToText.from_pretrained(
|
| 105 |
+
model_id,
|
| 106 |
+
torch_dtype="auto",
|
| 107 |
+
device_map="auto",
|
| 108 |
+
trust_remote_code=True,
|
| 109 |
+
)
|
| 110 |
+
```
|
| 111 |
+
|
| 112 |
+
For production serving, use vLLM, SGLang, or TokenSpeed with the **local directory** of the downloaded weights, not a guessed path. See Qwen's [vLLM recipe](https://recipes.vllm.ai/Qwen/Qwen3.8-27B) and [SGLang cookbook](https://docs.sglang.io/cookbook/autoregressive/Qwen/Qwen3.8-27B).
|
| 113 |
+
|
| 114 |
+
Thinking mode is on by default in the Qwen3.8 stack (`enable_thinking`, `preserve_thinking`, `reasoning_effort`). Keep those defaults unless you know you want instruct/non-thinking sampling.
|
| 115 |
+
|
| 116 |
+
## Evaluation
|
| 117 |
+
|
| 118 |
+
Preview: we do not publish a public leaderboard on this card yet. Cortex scores live training on a **private holdout**. Public splits are informational. General-capability checks run off the visible reward so they are not an overfitting target.
|
| 119 |
+
|
| 120 |
+
When a numbered release ships, this section will get frozen metrics and the eval recipe.
|
| 121 |
+
|
| 122 |
+
## Limitations
|
| 123 |
+
|
| 124 |
+
- Preview checkpoint: APIs, tokenizer extras, and chat template can still change.
|
| 125 |
+
- Post-training can move behavior away from the Qwen defaults (style, refusal, tool use). Compare against [`Qwen/Qwen3.8-27B`](https://huggingface.co/Qwen/Qwen3.8-27B) before you swap it in.
|
| 126 |
+
- Long-context and hour-scale video settings are inherited from Qwen; we have not re-tuned those knobs here.
|
| 127 |
+
- As with any open VLM, outputs can be wrong, biased, or unsafe. You are responsible for downstream filters.
|
| 128 |
+
|
| 129 |
+
## Files of ours vs Qwen's
|
| 130 |
+
|
| 131 |
+
| File | Source |
|
| 132 |
+
|---|---|
|
| 133 |
+
| Weights, tokenizer, `config.json`, processor configs | Forked from Qwen/Qwen3.8-27B |
|
| 134 |
+
| `LICENSE` | Apache 2.0 from the base (Copyright 2026 Alibaba Cloud) |
|
| 135 |
+
| `README.md`, `banner.jpg`, `NOTICE` | CortexLM |
|
| 136 |
+
|
| 137 |
+
## Citation
|
| 138 |
+
|
| 139 |
+
If you use Cortex-Mini-1-Preview, please cite both this work and Qwen3.8:
|
| 140 |
+
|
| 141 |
+
```bibtex
|
| 142 |
+
@misc{cortex-mini-1-preview,
|
| 143 |
+
title = {Cortex-Mini-1-Preview},
|
| 144 |
+
author = {CortexLM},
|
| 145 |
+
year = {2026},
|
| 146 |
+
url = {https://huggingface.co/CortexLM/Cortex-Mini-1-Preview}
|
| 147 |
+
}
|
| 148 |
+
|
| 149 |
+
@misc{qwen38,
|
| 150 |
+
title = {{Qwen3.8-Max}: A New Bar for Coding and Cowork},
|
| 151 |
+
url = {https://qwen.ai/blog?id=qwen3.8},
|
| 152 |
+
author = {{Qwen Team}},
|
| 153 |
+
month = {August},
|
| 154 |
+
year = {2026}
|
| 155 |
+
}
|
| 156 |
+
```
|
| 157 |
+
|
| 158 |
+
## Links
|
| 159 |
+
|
| 160 |
+
- Base model: [Qwen/Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B)
|
| 161 |
+
- Qwen collection: [Qwen3.8 on Hugging Face](https://huggingface.co/collections/Qwen/qwen38)
|
| 162 |
+
- CortexLM: [huggingface.co/CortexLM](https://huggingface.co/CortexLM)
|
| 163 |
+
- Cortex: [github.com/CortexLM/cortex](https://github.com/CortexLM/cortex)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|