From 3682c6a9794adeb8263a8f77afacd5ef6153dd1c Mon Sep 17 00:00:00 2001 From: Jin Qiao <89779290+JinBridger@users.noreply.github.com> Date: Thu, 13 Jun 2024 16:16:35 +0800 Subject: [PATCH] add glm4 and qwen2 to igpu perf (#11304) --- python/llm/test/benchmark/igpu-perf/1024-128.yaml | 1 + python/llm/test/benchmark/igpu-perf/1024-128_437.yaml | 1 + python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit.yaml | 1 + python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit_437.yaml | 1 + python/llm/test/benchmark/igpu-perf/2048-256.yaml | 1 + python/llm/test/benchmark/igpu-perf/2048-256_437.yaml | 1 + python/llm/test/benchmark/igpu-perf/32-32.yaml | 1 + python/llm/test/benchmark/igpu-perf/32-32_437.yaml | 1 + 8 files changed, 8 insertions(+) diff --git a/python/llm/test/benchmark/igpu-perf/1024-128.yaml b/python/llm/test/benchmark/igpu-perf/1024-128.yaml index 5584aba3..21757587 100644 --- a/python/llm/test/benchmark/igpu-perf/1024-128.yaml +++ b/python/llm/test/benchmark/igpu-perf/1024-128.yaml @@ -1,6 +1,7 @@ repo_id: - 'THUDM/chatglm3-6b' - 'THUDM/chatglm2-6b' + - 'THUDM/glm-4-9b-chat' - 'baichuan-inc/Baichuan2-7B-Chat' - 'baichuan-inc/Baichuan2-13B-Chat' - 'internlm/internlm-chat-7b' diff --git a/python/llm/test/benchmark/igpu-perf/1024-128_437.yaml b/python/llm/test/benchmark/igpu-perf/1024-128_437.yaml index 1b9cec75..16189637 100644 --- a/python/llm/test/benchmark/igpu-perf/1024-128_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/1024-128_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-7B-Instruct' - 'meta-llama/Meta-Llama-3-8B-Instruct' - '01-ai/Yi-6B-Chat' - 'microsoft/phi-2' diff --git a/python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit.yaml b/python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit.yaml index fee01274..12d5fa7e 100644 --- a/python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit.yaml +++ b/python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit.yaml @@ -1,6 +1,7 @@ repo_id: - 'THUDM/chatglm3-6b' - 'THUDM/chatglm2-6b' + - 'THUDM/glm-4-9b-chat' - 'baichuan-inc/Baichuan2-7B-Chat' - 'baichuan-inc/Baichuan2-13B-Chat' - 'internlm/internlm-chat-7b' diff --git a/python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit_437.yaml b/python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit_437.yaml index c74e988f..b9a29ddd 100644 --- a/python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/1024-128_loadlowbit_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-7B-Instruct' - 'meta-llama/Meta-Llama-3-8B-Instruct' - '01-ai/Yi-6B-Chat' - 'microsoft/phi-2' diff --git a/python/llm/test/benchmark/igpu-perf/2048-256.yaml b/python/llm/test/benchmark/igpu-perf/2048-256.yaml index 7e64f188..b995c294 100644 --- a/python/llm/test/benchmark/igpu-perf/2048-256.yaml +++ b/python/llm/test/benchmark/igpu-perf/2048-256.yaml @@ -1,6 +1,7 @@ repo_id: - 'THUDM/chatglm3-6b' - 'THUDM/chatglm2-6b' + - 'THUDM/glm-4-9b-chat' - 'baichuan-inc/Baichuan2-7B-Chat' - 'baichuan-inc/Baichuan2-13B-Chat' - 'internlm/internlm-chat-7b' diff --git a/python/llm/test/benchmark/igpu-perf/2048-256_437.yaml b/python/llm/test/benchmark/igpu-perf/2048-256_437.yaml index d6e971f9..c79f6616 100644 --- a/python/llm/test/benchmark/igpu-perf/2048-256_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/2048-256_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-7B-Instruct' - 'meta-llama/Meta-Llama-3-8B-Instruct' - '01-ai/Yi-6B-Chat' - 'microsoft/phi-2' diff --git a/python/llm/test/benchmark/igpu-perf/32-32.yaml b/python/llm/test/benchmark/igpu-perf/32-32.yaml index 20f6cb7b..9ac123ec 100644 --- a/python/llm/test/benchmark/igpu-perf/32-32.yaml +++ b/python/llm/test/benchmark/igpu-perf/32-32.yaml @@ -1,6 +1,7 @@ repo_id: - 'THUDM/chatglm3-6b' - 'THUDM/chatglm2-6b' + - 'THUDM/glm-4-9b-chat' - 'baichuan-inc/Baichuan2-7B-Chat' - 'baichuan-inc/Baichuan2-13B-Chat' - 'internlm/internlm-chat-7b' diff --git a/python/llm/test/benchmark/igpu-perf/32-32_437.yaml b/python/llm/test/benchmark/igpu-perf/32-32_437.yaml index 6b575cb4..13e70940 100644 --- a/python/llm/test/benchmark/igpu-perf/32-32_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/32-32_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-7B-Instruct' - 'meta-llama/Meta-Llama-3-8B-Instruct' - '01-ai/Yi-6B-Chat' - 'microsoft/phi-2'