Woosuk KwonandClaude Opus 5.5
8530aed2b7
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
71 lines
2.0 KiB
JSON
71 lines
2.0 KiB
JSON
{
|
|
"luccafong/deepseek_mtp_draft_random": {
|
|
"architectures": [
|
|
"DeepSeekMTPModel"
|
|
],
|
|
"model_type": "deepseek_mtp",
|
|
"text_model_type": "deepseek_mtp",
|
|
"hidden_size": 2560,
|
|
"total_num_hidden_layers": 1,
|
|
"total_num_attention_heads": 32,
|
|
"head_size": 576,
|
|
"vocab_size": 129280,
|
|
"total_num_kv_heads": 32,
|
|
"num_experts": 72,
|
|
"is_deepseek_mla": true,
|
|
"is_multimodal_model": false,
|
|
"dtype": "torch.bfloat16"
|
|
},
|
|
"eagle618/eagle-deepseek-v3-random": {
|
|
"architectures": [
|
|
"EagleDeepSeekMTPModel"
|
|
],
|
|
"model_type": "eagle",
|
|
"text_model_type": "eagle",
|
|
"hidden_size": 2560,
|
|
"total_num_hidden_layers": 1,
|
|
"total_num_attention_heads": 32,
|
|
"head_size": 576,
|
|
"vocab_size": 129280,
|
|
"total_num_kv_heads": 32,
|
|
"num_experts": 72,
|
|
"is_deepseek_mla": true,
|
|
"is_multimodal_model": false,
|
|
"dtype": "bfloat16"
|
|
},
|
|
"yuhuili/EAGLE-LLaMA3-Instruct-8B": {
|
|
"architectures": [
|
|
"EagleLlamaForCausalLM"
|
|
],
|
|
"model_type": "eagle",
|
|
"text_model_type": "eagle",
|
|
"hidden_size": 4096,
|
|
"total_num_hidden_layers": 1,
|
|
"total_num_attention_heads": 32,
|
|
"head_size": 128,
|
|
"vocab_size": 128256,
|
|
"total_num_kv_heads": 8,
|
|
"num_experts": 0,
|
|
"is_deepseek_mla": false,
|
|
"is_multimodal_model": false,
|
|
"dtype": "float16"
|
|
},
|
|
"yuhuili/EAGLE3-LLaMA3.1-Instruct-8B": {
|
|
"architectures": [
|
|
"Eagle3LlamaForCausalLM"
|
|
],
|
|
"model_type": "eagle",
|
|
"text_model_type": "eagle",
|
|
"hidden_size": 4096,
|
|
"total_num_hidden_layers": 1,
|
|
"total_num_attention_heads": 32,
|
|
"head_size": 128,
|
|
"vocab_size": 128256,
|
|
"total_num_kv_heads": 8,
|
|
"num_experts": 0,
|
|
"is_deepseek_mla": false,
|
|
"is_multimodal_model": false,
|
|
"dtype": "float16"
|
|
}
|
|
}
|