[Perf][GLM-5.2] Blackwell decode optimizations (#48597)
Signed-off-by: Peiyuan Zhou <peiyuanzhou1994@gmail.com> Signed-off-by: zhou <zhou@zhoudeMacBook-Pro.local> Signed-off-by: Stefan Koncarevic <stefan.koncarevic@amd.com> Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: Giancarlo Delfin <32987265+TheEpicDolphin@users.noreply.github.com> Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg> Co-authored-by: OpenAI Codex <codex@openai.com> Co-authored-by: zhou <zhou@zhoudeMacBook-Pro.local> Co-authored-by: stefankoncarevic <Stefan.Koncarevic@amd.com>
This commit is contained in:
co-authored by
Claude
Giancarlo Delfin
Thien Tran
OpenAI Codex
zhou
stefankoncarevic
parent
0d77325b10
commit
dd72658e7d
@@ -113,7 +113,7 @@ _TEXT_GENERATION_MODELS = {
|
||||
"Glm4ForCausalLM": ("glm4", "Glm4ForCausalLM"),
|
||||
"Glm4MoeForCausalLM": ("glm4_moe", "Glm4MoeForCausalLM"),
|
||||
"Glm4MoeLiteForCausalLM": ("glm4_moe_lite", "Glm4MoeLiteForCausalLM"),
|
||||
"GlmMoeDsaForCausalLM": ("deepseek_v2", "GlmMoeDsaForCausalLM"),
|
||||
"GlmMoeDsaForCausalLM": ("vllm.models.deepseek_v32", "GlmMoeDsaForCausalLM"),
|
||||
"GptOssForCausalLM": ("gpt_oss", "GptOssForCausalLM"),
|
||||
"GPT2LMHeadModel": ("gpt2", "GPT2LMHeadModel"),
|
||||
"GPTJForCausalLM": ("gpt_j", "GPTJForCausalLM"),
|
||||
@@ -628,6 +628,7 @@ _SPECULATIVE_DECODING_MODELS = {
|
||||
"Eagle3DeepseekV3ForCausalLM": ("deepseek_eagle3", "Eagle3DeepseekV2ForCausalLM"),
|
||||
"EagleDeepSeekMTPModel": ("deepseek_eagle", "EagleDeepseekV3ForCausalLM"),
|
||||
"DeepSeekMTPModel": ("deepseek_mtp", "DeepSeekMTP"),
|
||||
"DeepseekV32MTPModel": ("vllm.models.deepseek_v32", "DeepseekV32MTP"),
|
||||
"DeepSeekV4MTPModel": ("vllm.models.deepseek_v4", "DeepSeekV4MTP"),
|
||||
"MiniMaxM3MTP": ("vllm.models.minimax_m3", "MiniMaxM3MTP"),
|
||||
"BailingMoeV25MTPModel": ("bailing_moe_mtp", "BailingMoeV25MTPModel"),
|
||||
|
||||
Reference in New Issue
Block a user