Add Artificial Analysis evaluations for kimi-k2-thinking

README.md

@@ -1,3 +1,54 @@
----
-license: mit
----
+---
+license: mit
+model-index:
+- name: MackenzieTest
+  results:
+  - task:
+      type: evaluation
+    dataset:
+      name: Artificial Analysis Benchmarks
+      type: artificial_analysis
+    metrics:
+    - name: Artificial Analysis Intelligence Index
+      type: artificial_analysis_intelligence_index
+      value: 67
+    - name: Artificial Analysis Coding Index
+      type: artificial_analysis_coding_index
+      value: 52.2
+    - name: Artificial Analysis Math Index
+      type: artificial_analysis_math_index
+      value: 94.7
+    - name: Mmlu Pro
+      type: mmlu_pro
+      value: 0.848
+    - name: Gpqa
+      type: gpqa
+      value: 0.838
+    - name: Hle
+      type: hle
+      value: 0.223
+    - name: Livecodebench
+      type: livecodebench
+      value: 0.853
+    - name: Scicode
+      type: scicode
+      value: 0.424
+    - name: Aime 25
+      type: aime_25
+      value: 0.947
+    - name: Ifbench
+      type: ifbench
+      value: 0.681
+    - name: Lcr
+      type: lcr
+      value: 0.663
+    - name: Terminalbench Hard
+      type: terminalbench_hard
+      value: 0.291
+    - name: Tau2
+      type: tau2
+      value: 0.93
+    source:
+      name: Artificial Analysis API
+      url: https://artificialanalysis.ai
+---