Merge pull request #35 from Samoed/add_e5_instruct

EXTERNAL_MODEL_RESULTS.json

@@ -14202,6 +14202,703 @@
             ]
         }
     },
+    "multilingual-e5-large-instruct": {
+        "BitextMining": {
+            "f1": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "BornholmBitextMining": 55.42,
+                    "Tatoeba (rus-eng)": 93.7,
+                    "Tatoeba (spa-eng)": 98.53,
+                    "Tatoeba (isl-eng)": 95.2,
+                    "Tatoeba (jpn-eng)": 96.52,
+                    "Tatoeba (bre-eng)": 32.56,
+                    "Tatoeba (ast-eng)": 89.76,
+                    "Tatoeba (pms-eng)": 78.43,
+                    "Tatoeba (kab-eng)": 57.13,
+                    "Tatoeba (kat-eng)": 92.64,
+                    "Tatoeba (tur-eng)": 98.13,
+                    "Tatoeba (swh-eng)": 80.12,
+                    "Tatoeba (ukr-eng)": 94.8,
+                    "Tatoeba (epo-eng)": 98.38,
+                    "Tatoeba (lfn-eng)": 80.03,
+                    "Tatoeba (sqi-eng)": 96.45,
+                    "Tatoeba (cor-eng)": 11.54,
+                    "Tatoeba (pam-eng)": 19.75,
+                    "Tatoeba (gla-eng)": 77.87,
+                    "Tatoeba (amh-eng)": 87.64,
+                    "Tatoeba (jav-eng)": 85.29,
+                    "Tatoeba (lvs-eng)": 93.02,
+                    "Tatoeba (por-eng)": 94.91,
+                    "Tatoeba (arz-eng)": 82.26,
+                    "Tatoeba (aze-eng)": 93.89,
+                    "Tatoeba (yid-eng)": 90.47,
+                    "Tatoeba (gle-eng)": 84.37,
+                    "Tatoeba (pol-eng)": 97.33,
+                    "Tatoeba (ina-eng)": 96.1,
+                    "Tatoeba (ell-eng)": 94.85,
+                    "Tatoeba (fry-eng)": 80.83,
+                    "Tatoeba (kor-eng)": 91.77,
+                    "Tatoeba (ita-eng)": 94.37,
+                    "Tatoeba (ces-eng)": 96.23,
+                    "Tatoeba (fra-eng)": 94.97,
+                    "Tatoeba (ceb-eng)": 71.09,
+                    "Tatoeba (mal-eng)": 98.93,
+                    "Tatoeba (ido-eng)": 91.78,
+                    "Tatoeba (dtp-eng)": 14.1,
+                    "Tatoeba (kaz-eng)": 87.42,
+                    "Tatoeba (zsm-eng)": 96.62,
+                    "Tatoeba (dan-eng)": 95.87,
+                    "Tatoeba (nov-eng)": 83.42,
+                    "Tatoeba (tel-eng)": 95.51,
+                    "Tatoeba (arq-eng)": 64.1,
+                    "Tatoeba (dsb-eng)": 72.99,
+                    "Tatoeba (xho-eng)": 86.74,
+                    "Tatoeba (glg-eng)": 95.96,
+                    "Tatoeba (uig-eng)": 87.01,
+                    "Tatoeba (ang-eng)": 72.74,
+                    "Tatoeba (tzl-eng)": 59.17,
+                    "Tatoeba (deu-eng)": 99.33,
+                    "Tatoeba (ile-eng)": 89.94,
+                    "Tatoeba (lat-eng)": 74.17,
+                    "Tatoeba (eus-eng)": 84.67,
+                    "Tatoeba (bul-eng)": 94.52,
+                    "Tatoeba (hun-eng)": 94.52,
+                    "Tatoeba (mkd-eng)": 92.17,
+                    "Tatoeba (nds-eng)": 86.3,
+                    "Tatoeba (swg-eng)": 77.68,
+                    "Tatoeba (fao-eng)": 87.43,
+                    "Tatoeba (war-eng)": 73.83,
+                    "Tatoeba (csb-eng)": 68.19,
+                    "Tatoeba (afr-eng)": 95.67,
+                    "Tatoeba (bos-eng)": 95.86,
+                    "Tatoeba (cat-eng)": 95.53,
+                    "Tatoeba (hrv-eng)": 96.75,
+                    "Tatoeba (mhr-eng)": 16.98,
+                    "Tatoeba (tha-eng)": 96.59,
+                    "Tatoeba (tat-eng)": 85.7,
+                    "Tatoeba (oci-eng)": 72.35,
+                    "Tatoeba (urd-eng)": 93.22,
+                    "Tatoeba (kur-eng)": 77.3,
+                    "Tatoeba (lit-eng)": 91.71,
+                    "Tatoeba (cha-eng)": 50.55,
+                    "Tatoeba (ron-eng)": 97.32,
+                    "Tatoeba (pes-eng)": 94.92,
+                    "Tatoeba (cmn-eng)": 96.5,
+                    "Tatoeba (slv-eng)": 91.6,
+                    "Tatoeba (heb-eng)": 91.52,
+                    "Tatoeba (bel-eng)": 95.48,
+                    "Tatoeba (slk-eng)": 95.27,
+                    "Tatoeba (cbk-eng)": 83.24,
+                    "Tatoeba (yue-eng)": 92.84,
+                    "Tatoeba (hin-eng)": 97.47,
+                    "Tatoeba (nob-eng)": 98.33,
+                    "Tatoeba (gsw-eng)": 59.47,
+                    "Tatoeba (tgl-eng)": 96.0,
+                    "Tatoeba (nno-eng)": 94.06,
+                    "Tatoeba (est-eng)": 87.46,
+                    "Tatoeba (vie-eng)": 97.5,
+                    "Tatoeba (ber-eng)": 54.69,
+                    "Tatoeba (wuu-eng)": 92.4,
+                    "Tatoeba (tuk-eng)": 50.55,
+                    "Tatoeba (hye-eng)": 93.65,
+                    "Tatoeba (ara-eng)": 91.38,
+                    "Tatoeba (mon-eng)": 95.8,
+                    "Tatoeba (awa-eng)": 91.41,
+                    "Tatoeba (khm-eng)": 75.24,
+                    "Tatoeba (swe-eng)": 95.35,
+                    "Tatoeba (ind-eng)": 94.32,
+                    "Tatoeba (max-eng)": 74.56,
+                    "Tatoeba (hsb-eng)": 79.87,
+                    "Tatoeba (ben-eng)": 89.03,
+                    "Tatoeba (kzj-eng)": 14.96,
+                    "Tatoeba (uzb-eng)": 80.49,
+                    "Tatoeba (tam-eng)": 90.88,
+                    "Tatoeba (mar-eng)": 92.33,
+                    "Tatoeba (cym-eng)": 89.86,
+                    "Tatoeba (srp-eng)": 94.55,
+                    "Tatoeba (fin-eng)": 96.8,
+                    "Tatoeba (nld-eng)": 97.53,
+                    "Tatoeba (orv-eng)": 58.83
+                }
+            ]
+        },
+        "Classification": {
+            "accuracy": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "AllegroReviews": 52.43,
+                    "AmazonCounterfactualClassification (en-ext)": 66.42,
+                    "AmazonCounterfactualClassification (en)": 69.72,
+                    "AmazonCounterfactualClassification (de)": 65.61,
+                    "AmazonCounterfactualClassification (ja)": 70.47,
+                    "AmazonPolarityClassification": 96.24,
+                    "AmazonReviewsClassification (en)": 56.2,
+                    "AmazonReviewsClassification (de)": 54.54,
+                    "AmazonReviewsClassification (es)": 49.88,
+                    "AmazonReviewsClassification (fr)": 49.78,
+                    "AmazonReviewsClassification (ja)": 49.36,
+                    "AmazonReviewsClassification (zh)": 45.1,
+                    "AngryTweetsClassification": 59.53,
+                    "Banking77Classification": 78.04,
+                    "CBD": 74.21,
+                    "DanishPoliticalCommentsClassification": 33.07,
+                    "EmotionClassification": 54.99,
+                    "GeoreviewClassification": 55.9,
+                    "HeadlineClassification": 86.18,
+                    "IFlyTek": 44.06,
+                    "ImdbClassification": 94.61,
+                    "InappropriatenessClassification": 65.53,
+                    "JDReview": 80.23,
+                    "KinopoiskClassification": 66.12,
+                    "LccSentimentClassification": 60.27,
+                    "MTOPDomainClassification (en)": 91.18,
+                    "MTOPDomainClassification (de)": 90.03,
+                    "MTOPDomainClassification (es)": 89.1,
+                    "MTOPDomainClassification (fr)": 85.89,
+                    "MTOPDomainClassification (hi)": 86.29,
+                    "MTOPDomainClassification (th)": 83.35,
+                    "MTOPIntentClassification (en)": 68.05,
+                    "MTOPIntentClassification (de)": 68.55,
+                    "MTOPIntentClassification (es)": 69.86,
+                    "MTOPIntentClassification (fr)": 63.26,
+                    "MTOPIntentClassification (hi)": 63.38,
+                    "MTOPIntentClassification (th)": 64.8,
+                    "MasakhaNEWSClassification (amh)": 88.48,
+                    "MasakhaNEWSClassification (eng)": 82.12,
+                    "MasakhaNEWSClassification (fra)": 78.93,
+                    "MasakhaNEWSClassification (hau)": 80.94,
+                    "MasakhaNEWSClassification (ibo)": 77.28,
+                    "MasakhaNEWSClassification (lin)": 75.94,
+                    "MasakhaNEWSClassification (lug)": 70.36,
+                    "MasakhaNEWSClassification (orm)": 80.28,
+                    "MasakhaNEWSClassification (pcm)": 93.74,
+                    "MasakhaNEWSClassification (run)": 83.88,
+                    "MasakhaNEWSClassification (sna)": 87.91,
+                    "MasakhaNEWSClassification (som)": 68.61,
+                    "MasakhaNEWSClassification (swa)": 79.47,
+                    "MasakhaNEWSClassification (tir)": 72.87,
+                    "MasakhaNEWSClassification (xho)": 84.95,
+                    "MasakhaNEWSClassification (yor)": 82.99,
+                    "MassiveIntentClassification (sl)": 63.79,
+                    "MassiveIntentClassification (cy)": 56.22,
+                    "MassiveIntentClassification (sq)": 62.12,
+                    "MassiveIntentClassification (ar)": 55.87,
+                    "MassiveIntentClassification (vi)": 64.49,
+                    "MassiveIntentClassification (fr)": 66.88,
+                    "MassiveIntentClassification (sw)": 54.14,
+                    "MassiveIntentClassification (es)": 65.56,
+                    "MassiveIntentClassification (az)": 61.96,
+                    "MassiveIntentClassification (hy)": 61.73,
+                    "MassiveIntentClassification (ru)": 67.6,
+                    "MassiveIntentClassification (tr)": 66.3,
+                    "MassiveIntentClassification (ta)": 60.4,
+                    "MassiveIntentClassification (lv)": 61.49,
+                    "MassiveIntentClassification (jv)": 55.6,
+                    "MassiveIntentClassification (fi)": 66.02,
+                    "MassiveIntentClassification (fa)": 68.48,
+                    "MassiveIntentClassification (ka)": 50.74,
+                    "MassiveIntentClassification (ro)": 63.73,
+                    "MassiveIntentClassification (sv)": 68.76,
+                    "MassiveIntentClassification (ms)": 63.88,
+                    "MassiveIntentClassification (kn)": 58.98,
+                    "MassiveIntentClassification (ml)": 63.22,
+                    "MassiveIntentClassification (pt)": 67.28,
+                    "MassiveIntentClassification (my)": 57.06,
+                    "MassiveIntentClassification (tl)": 60.99,
+                    "MassiveIntentClassification (af)": 60.95,
+                    "MassiveIntentClassification (he)": 63.5,
+                    "MassiveIntentClassification (ja)": 68.29,
+                    "MassiveIntentClassification (zh-CN)": 67.6,
+                    "MassiveIntentClassification (km)": 46.88,
+                    "MassiveIntentClassification (mn)": 58.85,
+                    "MassiveIntentClassification (am)": 54.28,
+                    "MassiveIntentClassification (de)": 65.91,
+                    "MassiveIntentClassification (el)": 65.25,
+                    "MassiveIntentClassification (hu)": 65.54,
+                    "MassiveIntentClassification (en)": 70.89,
+                    "MassiveIntentClassification (id)": 65.45,
+                    "MassiveIntentClassification (ur)": 60.74,
+                    "MassiveIntentClassification (nl)": 68.18,
+                    "MassiveIntentClassification (da)": 65.08,
+                    "MassiveIntentClassification (th)": 62.58,
+                    "MassiveIntentClassification (pl)": 67.45,
+                    "MassiveIntentClassification (zh-TW)": 62.09,
+                    "MassiveIntentClassification (bn)": 61.73,
+                    "MassiveIntentClassification (is)": 57.29,
+                    "MassiveIntentClassification (te)": 61.38,
+                    "MassiveIntentClassification (ko)": 64.16,
+                    "MassiveIntentClassification (nb)": 65.85,
+                    "MassiveIntentClassification (it)": 66.82,
+                    "MassiveIntentClassification (hi)": 65.27,
+                    "MassiveScenarioClassification (ar)": 63.86,
+                    "MassiveScenarioClassification (id)": 70.52,
+                    "MassiveScenarioClassification (sl)": 68.6,
+                    "MassiveScenarioClassification (sq)": 69.09,
+                    "MassiveScenarioClassification (km)": 54.17,
+                    "MassiveScenarioClassification (fr)": 71.16,
+                    "MassiveScenarioClassification (bn)": 67.55,
+                    "MassiveScenarioClassification (am)": 61.43,
+                    "MassiveScenarioClassification (nl)": 73.21,
+                    "MassiveScenarioClassification (af)": 68.3,
+                    "MassiveScenarioClassification (fi)": 68.99,
+                    "MassiveScenarioClassification (ja)": 73.43,
+                    "MassiveScenarioClassification (en)": 73.88,
+                    "MassiveScenarioClassification (pt)": 69.76,
+                    "MassiveScenarioClassification (ka)": 59.88,
+                    "MassiveScenarioClassification (da)": 71.88,
+                    "MassiveScenarioClassification (jv)": 63.3,
+                    "MassiveScenarioClassification (zh-TW)": 68.04,
+                    "MassiveScenarioClassification (cy)": 62.98,
+                    "MassiveScenarioClassification (hu)": 71.72,
+                    "MassiveScenarioClassification (lv)": 67.54,
+                    "MassiveScenarioClassification (pl)": 71.44,
+                    "MassiveScenarioClassification (de)": 72.68,
+                    "MassiveScenarioClassification (ms)": 69.07,
+                    "MassiveScenarioClassification (ur)": 67.11,
+                    "MassiveScenarioClassification (kn)": 66.84,
+                    "MassiveScenarioClassification (ta)": 64.96,
+                    "MassiveScenarioClassification (is)": 67.14,
+                    "MassiveScenarioClassification (it)": 70.44,
+                    "MassiveScenarioClassification (sv)": 73.38,
+                    "MassiveScenarioClassification (te)": 67.53,
+                    "MassiveScenarioClassification (th)": 69.54,
+                    "MassiveScenarioClassification (fa)": 72.17,
+                    "MassiveScenarioClassification (mn)": 63.21,
+                    "MassiveScenarioClassification (ro)": 68.72,
+                    "MassiveScenarioClassification (ml)": 69.13,
+                    "MassiveScenarioClassification (az)": 66.18,
+                    "MassiveScenarioClassification (hi)": 69.83,
+                    "MassiveScenarioClassification (el)": 71.1,
+                    "MassiveScenarioClassification (nb)": 71.28,
+                    "MassiveScenarioClassification (tl)": 67.64,
+                    "MassiveScenarioClassification (he)": 67.63,
+                    "MassiveScenarioClassification (vi)": 69.17,
+                    "MassiveScenarioClassification (es)": 70.0,
+                    "MassiveScenarioClassification (ko)": 70.49,
+                    "MassiveScenarioClassification (tr)": 69.63,
+                    "MassiveScenarioClassification (ru)": 71.59,
+                    "MassiveScenarioClassification (zh-CN)": 72.41,
+                    "MassiveScenarioClassification (my)": 63.12,
+                    "MassiveScenarioClassification (sw)": 62.96,
+                    "MassiveScenarioClassification (hy)": 65.85,
+                    "MultilingualSentiment": 71.66,
+                    "NoRecClassification": 57.06,
+                    "NordicLangClassification": 76.57,
+                    "OnlineShopping": 91.83,
+                    "PAC": 65.68,
+                    "PolEmo2.0-IN": 80.96,
+                    "PolEmo2.0-OUT": 56.96,
+                    "RuReviewsClassification": 68.56,
+                    "RuSciBenchGRNTIClassification": 65.07,
+                    "RuSciBenchOECDClassification": 50.21,
+                    "TNews": 49.85,
+                    "ToxicConversationsClassification": 66.82,
+                    "TweetSentimentExtractionClassification": 59.18,
+                    "Waimai": 86.83
+                }
+            ]
+        },
+        "Clustering": {
+            "v_measure": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "GeoreviewClusteringP2P": 74.34,
+                    "MLSUMClusteringP2P (ru)": 57.77,
+                    "MLSUMClusteringS2S (ru)": 57.5,
+                    "MasakhaNEWSClusteringP2P (amh)": 74.82,
+                    "MasakhaNEWSClusteringP2P (eng)": 70.12,
+                    "MasakhaNEWSClusteringP2P (fra)": 70.48,
+                    "MasakhaNEWSClusteringP2P (hau)": 71.22,
+                    "MasakhaNEWSClusteringP2P (ibo)": 56.64,
+                    "MasakhaNEWSClusteringP2P (lin)": 72.94,
+                    "MasakhaNEWSClusteringP2P (lug)": 65.94,
+                    "MasakhaNEWSClusteringP2P (orm)": 45.85,
+                    "MasakhaNEWSClusteringP2P (pcm)": 88.57,
+                    "MasakhaNEWSClusteringP2P (run)": 64.64,
+                    "MasakhaNEWSClusteringP2P (sna)": 81.13,
+                    "MasakhaNEWSClusteringP2P (som)": 54.59,
+                    "MasakhaNEWSClusteringP2P (swa)": 40.55,
+                    "MasakhaNEWSClusteringP2P (tir)": 62.76,
+                    "MasakhaNEWSClusteringP2P (xho)": 57.61,
+                    "MasakhaNEWSClusteringP2P (yor)": 64.78,
+                    "MasakhaNEWSClusteringS2S (amh)": 67.38,
+                    "MasakhaNEWSClusteringS2S (eng)": 61.2,
+                    "MasakhaNEWSClusteringS2S (fra)": 69.48,
+                    "MasakhaNEWSClusteringS2S (hau)": 52.67,
+                    "MasakhaNEWSClusteringS2S (ibo)": 78.68,
+                    "MasakhaNEWSClusteringS2S (lin)": 80.26,
+                    "MasakhaNEWSClusteringS2S (lug)": 49.73,
+                    "MasakhaNEWSClusteringS2S (orm)": 44.39,
+                    "MasakhaNEWSClusteringS2S (pcm)": 74.0,
+                    "MasakhaNEWSClusteringS2S (run)": 64.34,
+                    "MasakhaNEWSClusteringS2S (sna)": 67.1,
+                    "MasakhaNEWSClusteringS2S (som)": 44.64,
+                    "MasakhaNEWSClusteringS2S (swa)": 42.09,
+                    "MasakhaNEWSClusteringS2S (tir)": 54.35,
+                    "MasakhaNEWSClusteringS2S (xho)": 43.19,
+                    "MasakhaNEWSClusteringS2S (yor)": 54.23,
+                    "RuSciBenchGRNTIClusteringP2P": 62.21,
+                    "RuSciBenchOECDClusteringP2P": 53.09,
+                    "TwentyNewsgroupsClustering": 51.03
+                }
+            ]
+        },
+        "PairClassification": {
+            "max_ap": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "CDSC-E": 76.17,
+                    "OpusparcusPC (de)": 97.56,
+                    "OpusparcusPC (en)": 98.91,
+                    "OpusparcusPC (fi)": 94.74,
+                    "OpusparcusPC (fr)": 94.73,
+                    "OpusparcusPC (ru)": 91.1,
+                    "OpusparcusPC (sv)": 95.78,
+                    "PSC": 99.31,
+                    "PawsXPairClassification (de)": 56.46,
+                    "PawsXPairClassification (en)": 64.4,
+                    "PawsXPairClassification (es)": 57.27,
+                    "PawsXPairClassification (fr)": 59.73,
+                    "PawsXPairClassification (ja)": 50.82,
+                    "PawsXPairClassification (ko)": 51.63,
+                    "PawsXPairClassification (zh)": 57.34,
+                    "SICK-E-PL": 79.69,
+                    "SprintDuplicateQuestions": 92.19,
+                    "TERRa": 63.77,
+                    "TwitterSemEval2015": 79.79,
+                    "TwitterURLCorpus": 86.73
+                },
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "CDSC-E": 76.17,
+                    "OpusparcusPC (de)": 97.56,
+                    "OpusparcusPC (en)": 98.92,
+                    "OpusparcusPC (fi)": 94.74,
+                    "OpusparcusPC (fr)": 94.73,
+                    "OpusparcusPC (ru)": 91.1,
+                    "OpusparcusPC (sv)": 95.78,
+                    "PSC": 99.31,
+                    "PawsXPairClassification (de)": 56.64,
+                    "PawsXPairClassification (en)": 64.43,
+                    "PawsXPairClassification (es)": 57.28,
+                    "PawsXPairClassification (fr)": 59.76,
+                    "PawsXPairClassification (ja)": 50.82,
+                    "PawsXPairClassification (ko)": 51.74,
+                    "PawsXPairClassification (zh)": 57.75,
+                    "SICK-E-PL": 79.69,
+                    "SprintDuplicateQuestions": 92.19,
+                    "TERRa": 63.89,
+                    "TwitterSemEval2015": 79.79,
+                    "TwitterURLCorpus": 86.73
+                }
+            ]
+        },
+        "Reranking": {
+            "map": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "AlloprofReranking": 74.68,
+                    "AskUbuntuDupQuestions": 64.41,
+                    "MMarcoReranking": 23.6,
+                    "MindSmallReranking": 33.07,
+                    "RuBQReranking": 75.84,
+                    "SciDocsRR": 85.75,
+                    "StackOverflowDupQuestions": 52.45,
+                    "SyntecReranking": 89.95,
+                    "T2Reranking": 67.12
+                },
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "MIRACLReranking (ru)": 62.49
+                }
+            ]
+        },
+        "Retrieval": {
+            "ndcg_at_10": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "AILACasedocs": 33.33,
+                    "AILAStatutes": 29.66,
+                    "ARCChallenge": 15.03,
+                    "AlloprofRetrieval": 52.12,
+                    "AlphaNLI": 24.87,
+                    "AppsRetrieval": 34.89,
+                    "ArguAna": 58.48,
+                    "ArguAna-PL": 44.6,
+                    "BSARDRetrieval": 24.61,
+                    "CmedqaRetrieval": 34.15,
+                    "CodeFeedbackMT": 39.87,
+                    "CodeFeedbackST": 75.89,
+                    "CodeSearchNetCCRetrieval (python)": 86.24,
+                    "CodeSearchNetCCRetrieval (javascript)": 79.2,
+                    "CodeSearchNetCCRetrieval (go)": 72.31,
+                    "CodeSearchNetCCRetrieval (ruby)": 81.9,
+                    "CodeSearchNetCCRetrieval (java)": 80.07,
+                    "CodeSearchNetCCRetrieval (php)": 72.49,
+                    "CodeSearchNetRetrieval (python)": 88.96,
+                    "CodeSearchNetRetrieval (javascript)": 75.37,
+                    "CodeSearchNetRetrieval (go)": 90.41,
+                    "CodeSearchNetRetrieval (ruby)": 79.56,
+                    "CodeSearchNetRetrieval (java)": 80.55,
+                    "CodeSearchNetRetrieval (php)": 83.7,
+                    "CodeTransOceanContest": 83.36,
+                    "CodeTransOceanDL": 29.01,
+                    "CosQA": 37.75,
+                    "CovidRetrieval": 75.8,
+                    "DuRetrieval": 85.16,
+                    "EcomRetrieval": 53.92,
+                    "FiQA-PL": 32.01,
+                    "FiQA2018": 48.42,
+                    "GerDaLIRSmall": 21.34,
+                    "HellaSwag": 32.02,
+                    "LEMBNarrativeQARetrieval": 26.71,
+                    "LEMBQMSumRetrieval": 26.08,
+                    "LEMBSummScreenFDRetrieval": 72.75,
+                    "LEMBWikimQARetrieval": 57.79,
+                    "LeCaRDv2": 64.22,
+                    "LegalBenchConsumerContractsQA": 76.8,
+                    "LegalBenchCorporateLobbying": 94.25,
+                    "LegalQuAD": 51.25,
+                    "LegalSummarization": 68.07,
+                    "MIRACLRetrieval (ru)": 66.08,
+                    "MMarcoRetrieval": 78.81,
+                    "MedicalRetrieval": 56.55,
+                    "MintakaRetrieval (ar)": 26.13,
+                    "MintakaRetrieval (de)": 35.13,
+                    "MintakaRetrieval (es)": 34.51,
+                    "MintakaRetrieval (fr)": 33.49,
+                    "MintakaRetrieval (hi)": 28.19,
+                    "MintakaRetrieval (it)": 35.01,
+                    "MintakaRetrieval (ja)": 26.31,
+                    "MintakaRetrieval (pt)": 35.11,
+                    "NFCorpus": 36.34,
+                    "NFCorpus-PL": 32.66,
+                    "PIQA": 32.26,
+                    "Quail": 8.63,
+                    "RARbCode": 71.22,
+                    "RARbMath": 71.95,
+                    "RiaNewsRetrieval": 83.26,
+                    "RuBQRetrieval": 73.9,
+                    "SCIDOCS": 19.24,
+                    "SCIDOCS-PL": 17.15,
+                    "SIQA": 7.33,
+                    "SciFact": 71.62,
+                    "SciFact-PL": 63.31,
+                    "SpartQA": 13.54,
+                    "StackOverflowQA": 85.82,
+                    "SyntecRetrieval": 87.8,
+                    "SyntheticText2SQL": 58.85,
+                    "T2Retrieval": 82.92,
+                    "TRECCOVID": 82.51,
+                    "TRECCOVID-PL": 62.03,
+                    "TempReasonL1": 1.2,
+                    "TempReasonL2Fact": 40.19,
+                    "TempReasonL2Pure": 3.64,
+                    "TempReasonL3Fact": 37.37,
+                    "TempReasonL3Pure": 9.82,
+                    "Touche2020": 27.4,
+                    "VideoRetrieval": 52.24,
+                    "WinoGrande": 54.27,
+                    "XPQARetrieval (ara-ara)": 48.56,
+                    "XPQARetrieval (eng-ara)": 34.01,
+                    "XPQARetrieval (ara-eng)": 45.13,
+                    "XPQARetrieval (deu-deu)": 79.18,
+                    "XPQARetrieval (eng-deu)": 50.55,
+                    "XPQARetrieval (deu-eng)": 75.23,
+                    "XPQARetrieval (spa-spa)": 65.07,
+                    "XPQARetrieval (eng-spa)": 41.75,
+                    "XPQARetrieval (spa-eng)": 61.02,
+                    "XPQARetrieval (fra-fra)": 72.72,
+                    "XPQARetrieval (eng-fra)": 46.17,
+                    "XPQARetrieval (fra-eng)": 69.33,
+                    "XPQARetrieval (hin-hin)": 76.62,
+                    "XPQARetrieval (eng-hin)": 18.12,
+                    "XPQARetrieval (hin-eng)": 71.95,
+                    "XPQARetrieval (ita-ita)": 77.03,
+                    "XPQARetrieval (eng-ita)": 41.42,
+                    "XPQARetrieval (ita-eng)": 70.6,
+                    "XPQARetrieval (jpn-jpn)": 74.36,
+                    "XPQARetrieval (eng-jpn)": 41.97,
+                    "XPQARetrieval (jpn-eng)": 70.5,
+                    "XPQARetrieval (kor-kor)": 39.74,
+                    "XPQARetrieval (eng-kor)": 34.9,
+                    "XPQARetrieval (kor-eng)": 37.39,
+                    "XPQARetrieval (pol-pol)": 52.53,
+                    "XPQARetrieval (eng-pol)": 34.22,
+                    "XPQARetrieval (pol-eng)": 48.53,
+                    "XPQARetrieval (por-por)": 51.17,
+                    "XPQARetrieval (eng-por)": 31.76,
+                    "XPQARetrieval (por-eng)": 48.79,
+                    "XPQARetrieval (tam-tam)": 49.69,
+                    "XPQARetrieval (eng-tam)": 6.8,
+                    "XPQARetrieval (tam-eng)": 40.83,
+                    "XPQARetrieval (cmn-cmn)": 66.42,
+                    "XPQARetrieval (eng-cmn)": 33.94,
+                    "XPQARetrieval (cmn-eng)": 59.78
+                }
+            ]
+        },
+        "STS": {
+            "cosine_spearman": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "AFQMC": 37.53,
+                    "ATEC": 43.27,
+                    "BIOSSES": 87.46,
+                    "BQ": 48.8,
+                    "CDSC-R": 92.35,
+                    "LCQMC": 76.06,
+                    "PAWSX": 15.06,
+                    "RUParaPhraserSTS": 75.4,
+                    "RuSTSBenchmarkSTS": 83.97,
+                    "SICK-R": 81.73,
+                    "SICK-R-PL": 77.62,
+                    "SICKFr": 80.2,
+                    "STS12": 82.53,
+                    "STS13": 88.05,
+                    "STS14": 84.83,
+                    "STS15": 91.02,
+                    "STS16": 87.32,
+                    "STS17 (ar-ar)": 82.71,
+                    "STS17 (nl-en)": 87.44,
+                    "STS17 (fr-en)": 86.28,
+                    "STS17 (ko-ko)": 84.31,
+                    "STS17 (es-es)": 89.4,
+                    "STS17 (en-ar)": 81.83,
+                    "STS17 (en-de)": 87.61,
+                    "STS17 (en-tr)": 80.47,
+                    "STS17 (es-en)": 87.03,
+                    "STS17 (en-en)": 90.33,
+                    "STS17 (it-en)": 88.25,
+                    "STS22 (es-en)": 78.65,
+                    "STS22 (de)": 61.53,
+                    "STS22 (zh)": 67.4,
+                    "STS22 (es)": 68.45,
+                    "STS22 (de-fr)": 65.52,
+                    "STS22 (pl)": 40.97,
+                    "STS22 (ru)": 65.17,
+                    "STS22 (en)": 68.67,
+                    "STS22 (fr)": 82.25,
+                    "STS22 (es-it)": 75.25,
+                    "STS22 (pl-en)": 74.86,
+                    "STS22 (fr-pl)": 61.98,
+                    "STS22 (ar)": 59.52,
+                    "STS22 (it)": 79.45,
+                    "STS22 (tr)": 67.2,
+                    "STS22 (zh-en)": 71.45,
+                    "STS22 (de-pl)": 54.6,
+                    "STS22 (de-en)": 58.22,
+                    "STSB": 81.68,
+                    "STSBenchmark": 88.39,
+                    "STSBenchmarkMultilingualSTS (de)": 85.38,
+                    "STSBenchmarkMultilingualSTS (nl)": 84.29,
+                    "STSBenchmarkMultilingualSTS (pl)": 83.42,
+                    "STSBenchmarkMultilingualSTS (fr)": 84.94,
+                    "STSBenchmarkMultilingualSTS (pt)": 84.9,
+                    "STSBenchmarkMultilingualSTS (it)": 84.38,
+                    "STSBenchmarkMultilingualSTS (es)": 86.15,
+                    "STSBenchmarkMultilingualSTS (zh)": 82.54,
+                    "STSBenchmarkMultilingualSTS (en)": 88.39,
+                    "STSBenchmarkMultilingualSTS (ru)": 83.86
+                },
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "AFQMC": 37.53,
+                    "ATEC": 43.27,
+                    "BIOSSES": 87.46,
+                    "BQ": 48.8,
+                    "CDSC-R": 92.35,
+                    "LCQMC": 76.06,
+                    "PAWSX": 15.06,
+                    "RUParaPhraserSTS": 75.4,
+                    "RuSTSBenchmarkSTS": 83.97,
+                    "SICK-R": 81.73,
+                    "SICK-R-PL": 77.62,
+                    "SICKFr": 80.2,
+                    "STS12": 82.53,
+                    "STS13": 88.05,
+                    "STS14": 84.83,
+                    "STS15": 91.02,
+                    "STS16": 87.32,
+                    "STS17 (ar-ar)": 82.71,
+                    "STS17 (nl-en)": 87.44,
+                    "STS17 (fr-en)": 86.28,
+                    "STS17 (ko-ko)": 84.31,
+                    "STS17 (es-es)": 89.4,
+                    "STS17 (en-ar)": 81.83,
+                    "STS17 (en-de)": 87.61,
+                    "STS17 (en-tr)": 80.47,
+                    "STS17 (es-en)": 87.03,
+                    "STS17 (en-en)": 90.33,
+                    "STS17 (it-en)": 88.25,
+                    "STS22 (es-en)": 78.65,
+                    "STS22 (de)": 61.53,
+                    "STS22 (zh)": 67.4,
+                    "STS22 (es)": 68.45,
+                    "STS22 (de-fr)": 65.52,
+                    "STS22 (pl)": 40.97,
+                    "STS22 (ru)": 65.17,
+                    "STS22 (en)": 68.67,
+                    "STS22 (fr)": 82.25,
+                    "STS22 (es-it)": 75.25,
+                    "STS22 (pl-en)": 74.86,
+                    "STS22 (fr-pl)": 61.98,
+                    "STS22 (ar)": 59.52,
+                    "STS22 (it)": 79.45,
+                    "STS22 (tr)": 67.2,
+                    "STS22 (zh-en)": 71.45,
+                    "STS22 (de-pl)": 54.6,
+                    "STS22 (de-en)": 58.22,
+                    "STSB": 81.68,
+                    "STSBenchmark": 88.39,
+                    "STSBenchmarkMultilingualSTS (de)": 85.38,
+                    "STSBenchmarkMultilingualSTS (nl)": 84.29,
+                    "STSBenchmarkMultilingualSTS (pl)": 83.42,
+                    "STSBenchmarkMultilingualSTS (fr)": 84.94,
+                    "STSBenchmarkMultilingualSTS (pt)": 84.9,
+                    "STSBenchmarkMultilingualSTS (it)": 84.38,
+                    "STSBenchmarkMultilingualSTS (es)": 86.15,
+                    "STSBenchmarkMultilingualSTS (zh)": 82.54,
+                    "STSBenchmarkMultilingualSTS (en)": 88.39,
+                    "STSBenchmarkMultilingualSTS (ru)": 83.86
+                }
+            ]
+        },
+        "Summarization": {
+            "cosine_spearman": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "SummEval": 30.46,
+                    "SummEvalFr": 31.72
+                },
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "SummEval": 30.46,
+                    "SummEvalFr": 31.72
+                }
+            ]
+        },
+        "MultilabelClassification": {
+            "accuracy": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "CEDRClassification": 50.01,
+                    "SensitiveTopicsClassification": 32.29
+                }
+            ]
+        },
+        "InstructionRetrieval": {
+            "p-MRR": [
+                {
+                    "Model": "<a target=\"_blank\" style=\"text-decoration: underline\" href=\"https://huggingface.co/intfloat/multilingual-e5-large-instruct\">multilingual-e5-large-instruct</a>",
+                    "Core17InstructionRetrieval": 1.82,
+                    "News21InstructionRetrieval": 1.5,
+                    "Robust04InstructionRetrieval": -4.52
+                }
+            ]
+        }
+    },
     "multilingual-e5-small": {
         "BitextMining": {
             "f1": [

model_meta.yaml

@@ -1072,6 +1072,15 @@ model_meta:
     is_proprietary: false
     is_sentence_transformers_compatible: true
     uses_instruct: false
+  multilingual-e5-large-instruct:
+    link: https://huggingface.co/intfloat/multilingual-e5-large-instruct
+    seq_len: 514
+    size: 560
+    dim: 1024
+    is_external: true
+    is_proprietary: false
+    is_sentence_transformers_compatible: true
+    uses_instruct: true
   multilingual-e5-small:
     link: https://huggingface.co/intfloat/multilingual-e5-small
     seq_len: 512