Upload folder using huggingface_hub

README.md

@@ -0,0 +1,44 @@
+---
+library_name: transformers
+license: mit
+---
+# sglang-EAGLE3-Qwen3-235B-A22B
+
+## Model Introduction
+The Eagle3 draft model was trained using the [SpecForge](https://github.com/sgl-project/SpecForge) framework for the Qwen3-235B-A22B model, leveraging a combination of UltraChat and ShareGPT datasets.
+
+
+## Benchmark Results
+gsm8k (200 questions)
+Accuracy: 0.965
+Invalid: 0.000
+Latency: 234.996 s
+Output throughput: 224.168 token/s
+<b>Accept length: 3.538</b>
+
+mtbench (80 questions)
+Number of questions: 80
+Output throughput: 241.5 token/s
+<b>Accept length: 3.019</b>
+
+
+## Usage
+You can use this Eagle3 draft model in [SGLang](https://github.com/sgl-project/sglang) with the following command:
+
+```bash
+python3 -m sglang.launch_server \
+    --model <Qwen3-235B-A22B> \
+    --speculative-algorithm EAGLE3 \
+    --speculative-draft-model-path <EAGLE3-Qwen3-235B-A22B> \
+    --speculative-num-steps 5 \
+    --speculative-eagle-topk 8 \
+    --speculative-num-draft-tokens 32 \
+    --mem-fraction-static 0.75 \
+    --tp 8 \
+    --enable-ep-moe \
+    --context-length 8192 \
+    --trust-remote-code \
+    --host 0.0.0.0 \
+    --port 30000 \
+    --dtype bfloat16
+```

config.json

@@ -0,0 +1,38 @@
+{
+  "architectures": [
+    "LlamaForCausalLMEagle3"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "draft_vocab_size": 32000,
+  "eagle_config": {
+    "eagle_aux_hidden_state_layer_ids": [
+      1,
+      46,
+      90
+    ],
+    "use_aux_hidden_state": true
+  },
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 24576,
+  "max_position_embeddings": 40960,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 1,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.53.2",
+  "use_cache": true,
+  "vocab_size": 151936
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c8158a7b35a0b51fbc2a569d7c15ad24f9c7f92ee80b874473abda7d2e1ce69
+size 2429992936