From df047f25ef9ad2c3bd049df34e5278d05557a278 Mon Sep 17 00:00:00 2001 From: Jinhe Tang Date: Wed, 7 Aug 2024 16:55:26 +0800 Subject: [PATCH 1/2] updated qwen1.5B to all transformer==4.37 yaml --- python/llm/test/benchmark/igpu-perf/1024-128_437.yaml | 1 + python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_437.yaml | 1 + python/llm/test/benchmark/igpu-perf/2048-256_int4_fp16_437.yaml | 1 + python/llm/test/benchmark/igpu-perf/3072-384_int4_fp16_437.yaml | 1 + python/llm/test/benchmark/igpu-perf/32-32_int4_fp16_437.yaml | 1 + python/llm/test/benchmark/igpu-perf/4096-512_int4_fp16_437.yaml | 1 + 6 files changed, 6 insertions(+) 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 f191801c7dc..b87e6c2c865 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/Qwen2-1.5B-Instruct' - 'Qwen/Qwen2-7B-Instruct' - 'microsoft/Phi-3-mini-4k-instruct' - 'microsoft/Phi-3-mini-128k-instruct' diff --git a/python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_437.yaml b/python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_437.yaml index f9db9131ca3..d7172186bb3 100644 --- a/python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-1.5B-Instruct' - 'Qwen/Qwen2-7B-Instruct' - 'microsoft/Phi-3-mini-4k-instruct' - 'microsoft/Phi-3-mini-128k-instruct' diff --git a/python/llm/test/benchmark/igpu-perf/2048-256_int4_fp16_437.yaml b/python/llm/test/benchmark/igpu-perf/2048-256_int4_fp16_437.yaml index fd4fbbfaec1..d6ee670ea29 100644 --- a/python/llm/test/benchmark/igpu-perf/2048-256_int4_fp16_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/2048-256_int4_fp16_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-1.5B-Instruct' - 'Qwen/Qwen2-7B-Instruct' - 'microsoft/Phi-3-mini-4k-instruct' - 'microsoft/Phi-3-mini-128k-instruct' diff --git a/python/llm/test/benchmark/igpu-perf/3072-384_int4_fp16_437.yaml b/python/llm/test/benchmark/igpu-perf/3072-384_int4_fp16_437.yaml index cfd7cc31afa..ddbf1cf4d19 100644 --- a/python/llm/test/benchmark/igpu-perf/3072-384_int4_fp16_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/3072-384_int4_fp16_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-1.5B-Instruct' - 'Qwen/Qwen2-7B-Instruct' - 'microsoft/Phi-3-mini-4k-instruct' - 'microsoft/Phi-3-mini-128k-instruct' diff --git a/python/llm/test/benchmark/igpu-perf/32-32_int4_fp16_437.yaml b/python/llm/test/benchmark/igpu-perf/32-32_int4_fp16_437.yaml index 93fdc926e5f..1a969165267 100644 --- a/python/llm/test/benchmark/igpu-perf/32-32_int4_fp16_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/32-32_int4_fp16_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-1.5B-Instruct' - 'Qwen/Qwen2-7B-Instruct' - 'microsoft/Phi-3-mini-4k-instruct' - 'microsoft/Phi-3-mini-128k-instruct' diff --git a/python/llm/test/benchmark/igpu-perf/4096-512_int4_fp16_437.yaml b/python/llm/test/benchmark/igpu-perf/4096-512_int4_fp16_437.yaml index 7c2632d3d96..c85da867aa2 100644 --- a/python/llm/test/benchmark/igpu-perf/4096-512_int4_fp16_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/4096-512_int4_fp16_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-1.5B-Instruct' - 'Qwen/Qwen2-7B-Instruct' - 'microsoft/Phi-3-mini-4k-instruct' - 'microsoft/Phi-3-mini-128k-instruct' From 6ca05fd8543a07205f415b1df72f76ce4063690b Mon Sep 17 00:00:00 2001 From: Jinhe Tang Date: Wed, 7 Aug 2024 16:56:44 +0800 Subject: [PATCH 2/2] updated qwen1.5B to all transformer==4.37 yaml --- .../benchmark/igpu-perf/1024-128_int4_fp16_loadlowbit_437.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_loadlowbit_437.yaml b/python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_loadlowbit_437.yaml index abd17aaa1e2..3839d0d2951 100644 --- a/python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_loadlowbit_437.yaml +++ b/python/llm/test/benchmark/igpu-perf/1024-128_int4_fp16_loadlowbit_437.yaml @@ -1,5 +1,6 @@ repo_id: - 'Qwen/Qwen1.5-7B-Chat' + - 'Qwen/Qwen2-1.5B-Instruct' - 'Qwen/Qwen2-7B-Instruct' - 'microsoft/Phi-3-mini-4k-instruct' - 'microsoft/Phi-3-mini-128k-instruct'