From bdd9e40c246fc41d24eed942c49d7d48702c2f95 Mon Sep 17 00:00:00 2001
From: nyd <113457465+nyedr@users.noreply.github.com>
Date: Fri, 13 Dec 2024 21:41:29 +0000
Subject: [PATCH] Added the recently released Llama 3.3 70B Instruct model
 information.

---
 models/meta/llama-3.3-70b-instruct/model.json | 95 +++++++++++++++++++
 1 file changed, 95 insertions(+)
 create mode 100644 models/meta/llama-3.3-70b-instruct/model.json

diff --git a/models/meta/llama-3.3-70b-instruct/model.json b/models/meta/llama-3.3-70b-instruct/model.json
new file mode 100644
index 0000000..f229f11
--- /dev/null
+++ b/models/meta/llama-3.3-70b-instruct/model.json
@@ -0,0 +1,95 @@
+{
+  "canonical_model_id": null,
+  "fine_tuned_from_model_id": null,
+  "name": "Llama 3.3 70B Instruct",
+  "description": "Llama 3.3 is a multilingual large language model optimized for dialogue use cases across multiple languages. It is a pretrained and instruction-tuned generative model with 70 billion parameters, outperforming many open-source and closed chat models on common industry benchmarks. Llama 3.3 supports a context length of 128,000 tokens and is designed for commercial and research use in multiple languages.",
+  "release_date": "2024-12-06",
+  "input_context_size": 128000,
+  "output_context_size": 128000,
+  "license": "Llama 3.3 Community License Agreement",
+  "multimodal": false,
+  "web_hydrated": false,
+  "knowledge_cutoff": "2023-12",
+  "api_ref_link": "https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct",
+  "playground_link": "https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct",
+  "paper_link": null,
+  "scorecard_blog_link": null,
+  "repo_link": "https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct",
+  "weights_link": "https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct",
+  "param_count": 70000000000,
+  "training_tokens": 15000000000000,
+  "qualitative_metrics": [
+    {
+      "dataset_name": "MMLU",
+      "score": 0.86,
+      "is_self_reported": true,
+      "analysis_method": "0-shot CoT",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    },
+    {
+      "dataset_name": "MMLU Pro",
+      "score": 0.689,
+      "is_self_reported": true,
+      "analysis_method": "0-shot CoT",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    },
+    {
+      "dataset_name": "IFEval",
+      "score": 0.921,
+      "is_self_reported": true,
+      "analysis_method": "Internal Evaluation",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    },
+    {
+      "dataset_name": "GPQA Diamond",
+      "score": 0.505,
+      "is_self_reported": true,
+      "analysis_method": "0-shot CoT",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    },
+    {
+      "dataset_name": "HumanEval",
+      "score": 0.884,
+      "is_self_reported": true,
+      "analysis_method": "Internal Evaluation",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    },
+    {
+      "dataset_name": "MBPP EvalPlus",
+      "score": 0.876,
+      "is_self_reported": true,
+      "analysis_method": "Internal Evaluation",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    },
+    {
+      "dataset_name": "MATH (CoT)",
+      "score": 0.77,
+      "is_self_reported": true,
+      "analysis_method": "0-shot CoT",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    },
+    {
+      "dataset_name": "BFCL v2",
+      "score": 0.773,
+      "is_self_reported": true,
+      "analysis_method": "Internal Evaluation",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    },
+    {
+      "dataset_name": "MGSM",
+      "score": 0.911,
+      "is_self_reported": true,
+      "analysis_method": "Internal Evaluation",
+      "date_recorded": "2024-12-06",
+      "source_link": "https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md"
+    }
+  ]
+}