Add support for the Llama 3.1 70B Instruct model to facilitate testing on high‑performance accelerators.

examples/MODULE.bazel

@@ -102,6 +102,30 @@ filegroup(
 )
 use_repo(huggingface, "Meta-Llama-3.1-8B-Instruct")
 
+huggingface.model(
+    name = "Meta-Llama-3.1-70B-Instruct",
+    build_file_content = """\
+package(default_visibility = ["//visibility:public"])
+filegroup(
+    name = "model",
+    srcs = glob(["*.safetensors"]) + ["model.safetensors.index.json"],
+)
+
+filegroup(
+    name = "tokenizer",
+    srcs = ["tokenizer.json"],
+)
+""",
+    commit = "945c8663693130f8be2ee66210e062158b2a9693",
+    includes = [
+        "*.safetensors",
+        "model.safetensors.index.json",
+        "tokenizer.json",
+    ],
+    model = "meta-llama/Meta-Llama-3.1-70B-Instruct",
+)
+use_repo(huggingface, "Meta-Llama-3.1-70B-Instruct")
+
 huggingface.model(
     name = "TinyLlama-1.1B-Chat-v1.0",
     build_file_content = """\

examples/llama/BUILD.bazel

@@ -34,6 +34,23 @@ native_binary(
     ],
 )
 
+native_binary(
+    name = "Llama-3.1-70B-Instruct",
+    src = ":llama",
+    args = [
+        "--model=$(location @Meta-Llama-3.1-70B-Instruct//:model.safetensors.index.json)",
+        "--tokenizer=$(location @Meta-Llama-3.1-70B-Instruct//:tokenizer)",
+        "--num-heads=64",
+        "--num-kv-heads=8",
+        "--rope-freq-base=500000",
+    ],
+    data = [
+        "@Meta-Llama-3.1-70B-Instruct//:model",
+        "@Meta-Llama-3.1-70B-Instruct//:model.safetensors.index.json",
+        "@Meta-Llama-3.1-70B-Instruct//:tokenizer",
+    ],
+)
+
 native_binary(
     name = "OpenLLaMA-3B",
     src = ":llama",