Safetensors
qwen2
linqq9 commited on
Commit
3f2a2c7
1 Parent(s): 1c6dba3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +27 -179
README.md CHANGED
@@ -343,19 +343,15 @@ The table below replicates and extends the format found in ["Granite-Function Ca
343
  </style>
344
  <table class="tg"><thead>
345
  <tr>
346
- <th class="tg-c1qe" rowspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Model</span></th>
347
- <th class="tg-c1qe" rowspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Size</span></th>
348
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">API-Bank</span></th>
349
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">API-Bank</span></th>
350
- <th class="tg-c1qe" colspan="2" rowspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Tool-Alpaca</span></th>
351
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Nexus</span></th>
352
- <th class="tg-c1qe" colspan="2" rowspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Average</span></th>
353
- </tr>
354
- <tr>
355
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">L-1</span></th>
356
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">L-2</span></th>
357
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Raven</span></th>
358
- </tr></thead>
359
  <tbody>
360
  <tr>
361
  <td class="tg-zor8"></td>
@@ -541,189 +537,41 @@ The table below replicates and extends the format found in ["Granite-Function Ca
541
  </style>
542
  <table class="tg"><thead>
543
  <tr>
544
- <th class="tg-c1qe" rowspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Model</span></th>
545
- <th class="tg-c1qe" rowspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Size</span></th>
546
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">API-Bank</span></th>
547
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">API-Bank</span></th>
548
- <th class="tg-c1qe" colspan="2" rowspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Tool-Alpaca</span></th>
549
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Nexus</span></th>
550
- <th class="tg-c1qe" colspan="2" rowspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Average</span></th>
551
- </tr>
552
- <tr>
553
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">L-1</span></th>
554
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">L-2</span></th>
555
- <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Raven</span></th>
556
  </tr></thead>
557
  <tbody>
558
  <tr>
559
- <td class="tg-zor8"></td>
560
- <td class="tg-zor8"></td>
561
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Func</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Name</span></td>
562
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Args</span></td>
563
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Func</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Name</span></td>
564
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Args</span></td>
565
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Func</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Name</span></td>
566
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Args</span></td>
567
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Func</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Name</span></td>
568
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Args</span></td>
569
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Func</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Name</span></td>
570
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Args</span></td>
571
- </tr>
572
- <tr>
573
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Functionary-small-v2.4</span></td>
574
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
575
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">78.00%</span></td>
576
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">70.00%</span></td>
577
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">54.00%</span></td>
578
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">45.00%</span></td>
579
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">88.00%</span></td>
580
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">47.00%</span></td>
581
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">82.00%</span></td>
582
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">64.00%</span></td>
583
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">75.50%</span></td>
584
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">56.50%</span></td>
585
  </tr>
586
  <tr>
587
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Gorilla-openfunctions-v2</span></td>
588
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
589
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">43.00%</span></td>
590
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">41.00%</span></td>
591
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">12.00%</span></td>
592
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">12.00%</span></td>
593
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">69.00%</span></td>
594
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">39.00%</span></td>
595
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">81.00%</span></td>
596
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">65.00%</span></td>
597
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">51.20%</span></td>
598
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">39.30%</span></td>
599
- </tr>
600
- <tr>
601
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Hermes-2-Pro-Mistral</span></td>
602
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
603
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">93.00%</span></td>
604
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">77.00%</span></td>
605
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">54.00%</span></td>
606
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">25.00%</span></td>
607
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">80.00%</span></td>
608
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">26.00%</span></td>
609
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">90.00%</span></td>
610
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">63.00%</span></td>
611
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">79.30%</span></td>
612
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">47.80%</span></td>
613
- </tr>
614
- <tr>
615
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Mistral-Instruct-v0.3</span></td>
616
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
617
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">79.00%</span></td>
618
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">69.00%</span></td>
619
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">69.00%</span></td>
620
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">46.00%</span></td>
621
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">33.00%</span></td>
622
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">33.00%</span></td>
623
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">71.00%</span></td>
624
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">54.00%</span></td>
625
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">63.00%</span></td>
626
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">50.50%</span></td>
627
- </tr>
628
- <tr>
629
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">CodeGemma-Instruct</span></td>
630
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
631
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">77.00%</span></td>
632
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">57.00%</span></td>
633
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">59.00%</span></td>
634
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">38.00%</span></td>
635
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">59.00%</span></td>
636
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">31.00%</span></td>
637
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">84.00%</span></td>
638
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">68.00%</span></td>
639
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">69.80%</span></td>
640
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">48.50%</span></td>
641
- </tr>
642
- <tr>
643
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Nexusflow-Raven-v2</span></td>
644
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">13B</span></td>
645
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">51.00%</span></td>
646
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">42.00%</span></td>
647
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">28.00%</span></td>
648
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">22.00%</span></td>
649
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">85.00%</span></td>
650
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">37.00%</span></td>
651
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">92.00%</span></td>
652
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">75.00%</span></td>
653
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">64.00%</span></td>
654
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">44.00%</span></td>
655
- </tr>
656
- <tr>
657
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">C4AI-Command-R-v01</span></td>
658
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">35B</span></td>
659
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">93.00%</span></td>
660
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">76.00%</span></td>
661
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">77.00%</span></td>
662
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">54.00%</span></td>
663
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">90.00%</span></td>
664
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">42.00%</span></td>
665
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">93.00%</span></td>
666
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">71.00%</span></td>
667
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">88.30%</span></td>
668
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">60.80%</span></td>
669
- </tr>
670
- <tr>
671
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Meta-Llama-3-70B-Instruct</span></td>
672
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">70B</span></td>
673
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">85.00%</span></td>
674
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">67.00%</span></td>
675
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">69.00%</span></td>
676
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">52.00%</span></td>
677
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">78.00%</span></td>
678
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">43.00%</span></td>
679
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">70.00%</span></td>
680
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">52.00%</span></td>
681
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">75.50%</span></td>
682
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">53.50%</span></td>
683
  </tr>
684
  <tr>
685
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">GRANITE-20B-FUNCTIONCALLING</span></td>
686
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">20B</span></td>
687
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">91.00%</span></td>
688
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">71.00%</span></td>
689
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">83.00%</span></td>
690
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">60.00%</span></td>
691
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">89.00%</span></td>
692
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">44.00%</span></td>
693
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">92.00%</span></td>
694
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">72.00%</span></td>
695
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">88.80%</span></td>
696
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">61.80%</span></td>
697
  </tr>
698
  <tr>
699
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">xlam-7b-fc-r</span></td>
700
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
701
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">90.00%</span></td>
702
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">80.70%</span></td>
703
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">68.90%</span></td>
704
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">60.70%</span></td>
705
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">67.30%</span></td>
706
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">59.00%</span></td>
707
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">54.10%</span></td>
708
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">57.50%</span></td>
709
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">70.10%</span></td>
710
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">64.50%</span></td>
711
  </tr>
712
  <tr>
713
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Hammer-7b</span></td>
714
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
715
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">93.80%</span></td>
716
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">85.90%</span></td>
717
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">79.20%</span></td>
718
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">64.40%</span></td>
719
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">82.30%</span></td>
720
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">59.90%</span></td>
721
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">92.50%</span></td>
722
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">77.40%</span></td>
723
- <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">86.90%</span></td>
724
- <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">71.90%</span></td>
725
  </tr>
726
-
727
  </tbody></table>
728
 
729
  ## Upcoming Developments
@@ -740,7 +588,7 @@ import torch
740
  from transformers import AutoModelForCausalLM, AutoTokenizer
741
 
742
 
743
- model_name = "/home/notebook/data/group/ComplexTaskDecision/Hammer/ckpt/select_caller/xlam_7B/xlam_mask3_0.33_hammer_qwen7b_batch32/merge_step4220_bf16"
744
  model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto", trust_remote_code=True)
745
  tokenizer = AutoTokenizer.from_pretrained(model_name)
746
 
@@ -812,7 +660,7 @@ get_stock_price={
812
  }
813
  }
814
 
815
- # Helper function to convert openai format tools to our more concise xLAM format
816
  def convert_to_format_tool(tools):
817
  ''''''
818
  if isinstance(tools, dict):
 
343
  </style>
344
  <table class="tg"><thead>
345
  <tr>
346
+ <th class="tg-c1qe"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Model</span></th>
347
+ <th class="tg-c1qe"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Size</span></th>
348
+ <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">API-Bank</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">L-1</span></th>
349
+ <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">API-Bank</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">L-2</span></th>
350
+ <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Tool-Alpaca</span></th>
351
+ <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Nexus</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Raven</span></th>
352
+ <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Average</span></th>
353
+ </tr>
354
+ </thead>
 
 
 
 
355
  <tbody>
356
  <tr>
357
  <td class="tg-zor8"></td>
 
537
  </style>
538
  <table class="tg"><thead>
539
  <tr>
540
+ <th class="tg-c1qe"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Model</span></th>
541
+ <th class="tg-c1qe"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Size</span></th>
542
+ <th class="tg-c1qe" colspan="2"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">SealTool(Single-Tool)</span></th>
 
 
 
 
 
 
 
 
 
543
  </tr></thead>
544
  <tbody>
545
  <tr>
546
+ <td class="tg-c1qe"></td>
547
+ <td class="tg-c1qe"></td>
548
+ <td class="tg-c1qe"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Func</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Name</span></td>
549
+ <td class="tg-c1qe"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">F1</span> <span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#B9C9FE">Args</span></td>
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
550
  </tr>
551
  <tr>
552
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">Gorilla-openfunctions-v2</span></td>
553
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
554
+ <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">93.20%</span></td>
555
+ <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">91.10%</span></td>
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
556
  </tr>
557
  <tr>
558
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">GRANITE-20B-FUNCTIONCALLING</span></td>
559
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">20B</span></td>
560
+ <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">94.90%</span></td>
561
+ <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">92.70%</span></td>
 
 
 
 
 
 
 
 
562
  </tr>
563
  <tr>
564
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">xlam-7b-fc-r</span></td>
565
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
566
+ <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">79.00%</span></td>
567
+ <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">76.90%</span></td>
 
 
 
 
 
 
 
 
568
  </tr>
569
  <tr>
570
+ <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">hammer-7b</span></td>
571
  <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">7B</span></td>
572
+ <td class="tg-roal"><span style="font-weight:700;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">97.40%</span></td>
573
+ <td class="tg-zor8"><span style="font-weight:400;font-style:normal;text-decoration:none;color:#000;background-color:#E8EDFF">91.70%</span></td>
 
 
 
 
 
 
 
 
574
  </tr>
 
575
  </tbody></table>
576
 
577
  ## Upcoming Developments
 
588
  from transformers import AutoModelForCausalLM, AutoTokenizer
589
 
590
 
591
+ model_name = "MadeAgents/Hammer-7b"
592
  model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto", trust_remote_code=True)
593
  tokenizer = AutoTokenizer.from_pretrained(model_name)
594
 
 
660
  }
661
  }
662
 
663
+
664
  def convert_to_format_tool(tools):
665
  ''''''
666
  if isinstance(tools, dict):