Synchronizing local compiler cache.

.gitattributes

@@ -11894,3 +11894,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_14928681760864605802+f7f529f3/model.neff
 neuronxcc-2.21.18209.0+043b1bf7/MODULE_2092161936687880602+f7f529f3/model.neff filter=lfs diff=lfs merge=lfs -text
 neuronxcc-2.21.18209.0+043b1bf7/MODULE_2927013901654163041+f7f529f3/model.neff filter=lfs diff=lfs merge=lfs -text
 neuronxcc-2.21.18209.0+043b1bf7/MODULE_2962232050317550424+f7f529f3/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.21.18209.0+043b1bf7/MODULE_669f33fc567910d306ab+564b7b5b/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.21.18209.0+043b1bf7/MODULE_b85364df3fc31288dc1c+d0d57c8a/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.21.18209.0+043b1bf7/MODULE_b85364df3fc31288dc1c+d0d57c8a/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text

neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3_moe/Qwen/Qwen3-Coder-30B-A3B-Instruct/1215d01d0b52e0c80683.json

@@ -0,0 +1,68 @@
+{
+  "_entry_class": "SingleModelCacheEntry",
+  "_model_id": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
+  "_task": "text-generation",
+  "architectures": [
+    "Qwen3MoeForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "decoder_sparse_step": 1,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5472,
+  "max_position_embeddings": 262144,
+  "max_window_layers": 28,
+  "mlp_only_layers": [],
+  "model_type": "qwen3_moe",
+  "moe_intermediate_size": 768,
+  "neuron": {
+    "_serialized_key": "NxDNeuronConfig",
+    "batch_size": 16,
+    "capacity_factor": null,
+    "checkpoint_id": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
+    "checkpoint_revision": "573fa3901e5799703b1e60825b0ec024a4c0f1d3",
+    "continuous_batching": true,
+    "enable_bucketing": false,
+    "ep_degree": 1,
+    "fused_qkv": false,
+    "glu_mlp": true,
+    "local_ranks_size": 4,
+    "max_batch_size": 16,
+    "max_context_length": 4096,
+    "max_topk": 256,
+    "n_active_tokens": 4096,
+    "neuronxcc_version": "2.21.18209.0+043b1bf7",
+    "on_device_sampling": true,
+    "optimum_neuron_version": "0.4.1.dev0",
+    "output_logits": false,
+    "pp_degree": 1,
+    "sequence_length": 4096,
+    "speculation_length": 0,
+    "start_rank_id": 0,
+    "target": "trn2",
+    "torch_dtype": "bfloat16",
+    "tp_degree": 4
+  },
+  "norm_topk_prob": true,
+  "num_attention_heads": 32,
+  "num_experts": 128,
+  "num_experts_per_tok": 8,
+  "num_hidden_layers": 48,
+  "num_key_value_heads": 4,
+  "output_router_logits": false,
+  "qkv_bias": false,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 10000000,
+  "router_aux_loss_coef": 0.0,
+  "shared_expert_intermediate_size": 0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "use_cache": true,
+  "use_qk_norm": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3d222b1e460d753301+d0d57c8a/compile_flags.json

@@ -0,0 +1 @@
+["--target=trn2", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]

neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3d222b1e460d753301+d0d57c8a/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e0800ecd46f0d680b0c9bf5514c1b241a43ff41648d510d053ed27d3d3b0fa8
+size 14468174

neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3d222b1e460d753301+d0d57c8a/model.log

@@ -0,0 +1,2 @@
+Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_2a3d222b1e460d753301+d0d57c8a.hlo_module.pb', '--output', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_2a3d222b1e460d753301+d0d57c8a.neff', '--target=trn2', '--enable-saturate-infinity', '--enable-mixed-precision-accumulation', '--model-type', 'transformer', '-O1', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2', '--auto-cast=none', '--internal-enable-dge-levels', 'vector_dynamic_offsets', '--internal-hlo2tensorizer-options=--verify-hlo=true', '--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt', '--enable-internal-neff-wrapper', '--verbose=35']: 2025-10-16T11:15:30Z 2025-10-16 11:15:30.307267: F hilo/hlo_passes/NeuronHloVerifier.cc:504] [ERROR] [NCC_VRF009] Memory requirement exceeds target architecture's HBM limit. Needed 18507002752 bytes (17 GB) vs. available 17179869184 bytes (16 GB). TIP: Consider using smaller batches or applying model parallelism
+

neuronxcc-2.21.18209.0+043b1bf7/MODULE_669f33fc567910d306ab+564b7b5b/compile_flags.json

@@ -0,0 +1 @@
+["--target=trn2", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]

neuronxcc-2.21.18209.0+043b1bf7/MODULE_669f33fc567910d306ab+564b7b5b/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f8551db25d6c000741567bc5ef73f556581b39bb73b58a4b4580ed042cb0023
+size 10995696

neuronxcc-2.21.18209.0+043b1bf7/MODULE_669f33fc567910d306ab+564b7b5b/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc379dc0f4e63d796eced8a100b901e4de0141de37d9cb52834aacb00835921a
+size 16159744

neuronxcc-2.21.18209.0+043b1bf7/MODULE_b85364df3fc31288dc1c+d0d57c8a/compile_flags.json

@@ -0,0 +1 @@
+["--target=trn2", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]

neuronxcc-2.21.18209.0+043b1bf7/MODULE_b85364df3fc31288dc1c+d0d57c8a/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abcac81baad6aef1db291146b7601567b89bcb5db34f7676b18942692a0de687
+size 12051954

neuronxcc-2.21.18209.0+043b1bf7/MODULE_b85364df3fc31288dc1c+d0d57c8a/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3a41cf2fa4c52126a3d319bd14e3d84c3a8a2d8c415b52f94eb7bd3c607ccef
+size 3984384

neuronxcc-2.21.18209.0+043b1bf7/MODULE_b85364df3fc31288dc1c+d0d57c8a/wrapped_neff.hlo

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7bdbd9d86286b4d2d40a110747321080624d5ea42200084e2278c84abf095d8
+size 4257451