Add pr tests (#31952)
Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: sglang-npu-bot <sglangnpu@163.com> Co-authored-by: Cherry_ming <136634645@qq.com>
This commit is contained in:
co-authored by
Claude
sglang-npu-bot
Cherry_ming
parent
fd96a35fb0
commit
33ecf4bcd8
@@ -114,8 +114,11 @@ jobs:
|
||||
test_config:
|
||||
- name: qwen3_32b_w8a8_2p_in3k5_out1k5_50ms_a2
|
||||
runner: linux-aarch64-a2-4
|
||||
test_case: test/registered/ascend/performance/qwen3_32b/test_npu_qwen3_32b_w8a8_2p_in3k5_out1k5_50ms_a2.py
|
||||
test_case: test/registered/npu/performance/qwen3_32b/test_npu_qwen3_32b_w8a8_2p_in3k5_out1k5_50ms_a2.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_32b_w8a8_2p_in3k5_out1k5_50ms_gpqa_a2
|
||||
runner: linux-aarch64-a2-4
|
||||
test_case: test/registered/npu/accuracy/qwen3_32b/test_npu_qwen3_32b_w8a8_2p_in3k5_out1k5_50ms_gpqa_a2.py
|
||||
uses: ./.github/workflows/nightly-test-npu-e2e-single-node.yml
|
||||
with:
|
||||
runner: ${{ matrix.test_config.runner }}
|
||||
@@ -138,119 +141,158 @@ jobs:
|
||||
# qwen3_6_35b_a3b performance tests
|
||||
- name: qwen3_6_35b_a3b_1p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in3k5_out1k5_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_35b_a3b_1p_aime26
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/accuracy/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_aime26.py
|
||||
test_case: test/registered/npu/accuracy/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_aime26.py
|
||||
- name: qwen3_6_35b_a3b_1p_in64k_out1k_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in64k_out1k_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in64k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_35b_a3b_1p_in128k_out1k_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in128k_out1k_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in128k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_35b_a3b_1p_in64k_out1k_prefix90_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in64k_out1k_prefix90_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_35b_a3b_1p_in64k_out1k_prefix90_50ms_aime26
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in64k_out1k_prefix90_50ms_aime26.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in64k_out1k_prefix90_50ms_aime26.py
|
||||
- name: qwen3_6_35b_a3b_1p_in128k_out1k_prefix90_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in128k_out1k_prefix90_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in128k_out1k_prefix90_50ms.py
|
||||
test_type: 'perf'
|
||||
# qwen3_6_27b performance tests
|
||||
- name: qwen3_6_27b_w8a8_1p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_27b_w8a8_1p_in3k5_out1k5_50ms_gpqa
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in3k5_out1k5_50ms_gpqa.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in3k5_out1k5_50ms_gpqa.py
|
||||
- name: qwen3_6_27b_w8a8_2p_in16k_out1k_50ms
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_2p_in16k_out1k_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_2p_in16k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_27b_w8a8_1p_in64k_out1k_50ms
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in64k_out1k_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in64k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_27b_w8a8_2p_in128k_out1k_50ms
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_2p_in128k_out1k_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_2p_in128k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_27b_2p_in64k_out1k_prefix90_50ms
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_2p_in64k_out1k_prefix90_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_2p_in64k_out1k_prefix90_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_27b_1p_in1024x1024_30_out1024_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_1p_in1024x1024_30_out1024_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_1p_in1024x1024_30_out1024_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_27b_1p_in1080p_30_out256_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_1p_in1080p_30_out256_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_1p_in1080p_30_out256_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_6_27b_1p_gpqa
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/accuracy/qwen3_6_27b/test_npu_qwen3_6_27b_1p_gpqa.py
|
||||
test_case: test/registered/npu/accuracy/qwen3_6_27b/test_npu_qwen3_6_27b_1p_gpqa.py
|
||||
# qwen3_32b performance tests
|
||||
- name: qwen3_32b_w8a8_2p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/npu/performance/qwen3_32b/test_npu_qwen3_32b_w8a8_2p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_32b_w8a8_2p_in3k5_out1k5_50ms_gpqa
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/ascend/performance/qwen3_32b/test_npu_qwen3_32b_w8a8_2p_in3k5_out1k5_50ms_gpqa.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/qwen3_32b/test_npu_qwen3_32b_w8a8_2p_in3k5_out1k5_50ms_gpqa.py
|
||||
- name: qwen3_32b_bf16_8p_in18k_out4k_6ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/ascend/performance/qwen3_32b/test_npu_qwen3_32b_bf16_8p_in18k_out4k_6ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_32b/test_npu_qwen3_32b_bf16_8p_in18k_out4k_6ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_32b_bf16_8p_gpqa
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/ascend/accuracy/qwen3_32b/test_npu_qwen3_32b_bf16_8p_gpqa.py
|
||||
test_case: test/registered/npu/accuracy/qwen3_32b/test_npu_qwen3_32b_bf16_8p_gpqa.py
|
||||
# qwen3_30b_a3b performance tests
|
||||
- name: qwen3_30b_w8a8_1p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3_30b_a3b/test_npu_qwen3_30b_w8a8_1p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_30b_w8a8_1p_in3k5_out1k5_50ms_aime25
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_30b_a3b/test_npu_qwen3_30b_w8a8_1p_in3k5_out1k5_50ms_aime25.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/qwen3_30b_a3b/test_npu_qwen3_30b_w8a8_1p_in3k5_out1k5_50ms_aime25.py
|
||||
# qwen3-8b performance tests
|
||||
- name: qwen3_8b_w8a8_1p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3-8b/test_npu_qwen3_8b_w8a8_1p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_8b_w8a8_1p_in3k5_out1k5_50ms_gpqa
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3-8b/test_npu_qwen3_8b_w8a8_1p_in3k5_out1k5_50ms_gpqa.py
|
||||
test_case: test/registered/npu/accuracy/qwen3-8b/test_npu_qwen3_8b_w8a8_1p_in3k5_out1k5_50ms_gpqa.py
|
||||
- name: qwen3_8b_w8a8_1p_in6k_out1k5_bs16
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3-8b/test_npu_qwen3_8b_w8a8_1p_in6k_out1k5_bs16.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_8b_w8a8_1p_in6k_out1k5_bs16_gpqa
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3-8b/test_npu_qwen3_8b_w8a8_1p_in6k_out1k5_bs16_gpqa.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/qwen3-8b/test_npu_qwen3_8b_w8a8_1p_in6k_out1k5_bs16_gpqa.py
|
||||
# qwen3_next_80b_a3b_instruct performance tests
|
||||
- name: qwen3_next_80b_w8a8_2p_in6k_out1k5_bs16
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/npu/performance/qwen3_next_80b_a3b_instruct/test_npu_qwen3_next_80b_w8a8_2p_in6k_out1k5_bs16.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_next_80b_w8a8_2p_in6k_out1k5_bs16_aime25
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/ascend/performance/qwen3_next_80b_a3b_instruct/test_npu_qwen3_next_80b_w8a8_2p_in6k_out1k5_bs16_aime25.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/qwen3_next_80b_a3b_instruct/test_npu_qwen3_next_80b_w8a8_2p_in6k_out1k5_bs16_aime25.py
|
||||
# minimax_m2_5 performance tests
|
||||
- name: minimax_m2_5_w8a8_8p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/minimax_m2_5/test_npu_minimax_m2_5_w8a8_8p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: minimax_m2_5_w8a8_8p_in3k5_out1k5_50ms_gpqa
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/ascend/performance/minimax_m2_5/test_npu_minimax_m2_5_w8a8_8p_in3k5_out1k5_50ms_gpqa.py
|
||||
test_case: test/registered/npu/accuracy/minimax_m2_5/test_npu_minimax_m2_5_w8a8_8p_in3k5_out1k5_50ms_gpqa.py
|
||||
- name: minimax_m2_5_w8a8_4p_in64k_out1k_prefix90_50ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/minimax_m2_5/test_npu_minimax_m2_5_w8a8_4p_in64k_out1k_prefix90_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: minimax_m2_5_w8a8_4p_in64k_out1k_prefix90_50ms_gpqa
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/ascend/performance/minimax_m2_5/test_npu_minimax_m2_5_w8a8_4p_in64k_out1k_prefix90_50ms_gpqa.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/minimax_m2_5/test_npu_minimax_m2_5_w8a8_4p_in64k_out1k_prefix90_50ms_gpqa.py
|
||||
# deepseek_v3_2 accuracy tests
|
||||
- name: deepseek_v3_2_8p_aime25
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/ascend/accuracy/deepseek_v3_2/test_npu_deepseek_v3_2_8p_aime25.py
|
||||
test_case: test/registered/npu/accuracy/deepseek_v3_2/test_npu_deepseek_v3_2_8p_aime25.py
|
||||
# glm4_7_flash accuracy tests
|
||||
- name: glm4_7_flash_1p_aime25
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/accuracy/glm4_7_flash/test_npu_glm4_7_flash_1p_aime25.py
|
||||
test_case: test/registered/npu/accuracy/glm4_7_flash/test_npu_glm4_7_flash_1p_aime25.py
|
||||
# qwen3_vl_8b_thinking accuracy tests
|
||||
- name: qwen3_vl_8b_thinking_1p_mmmu
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/accuracy/qwen3_vl_8b_thinking/test_npu_qwen3_vl_8b_thinking_1p_mmmu.py
|
||||
test_case: test/registered/npu/accuracy/qwen3_vl_8b_thinking/test_npu_qwen3_vl_8b_thinking_1p_mmmu.py
|
||||
# qwen3_vl_30b_a3b_thinking accuracy tests
|
||||
- name: qwen3_vl_30b_a3b_thinking_1p_mmmu
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/accuracy/qwen3_vl_30b_a3b_thinking/test_npu_qwen3_vl_30b_a3b_thinking_1p_mmmu.py
|
||||
test_case: test/registered/npu/accuracy/qwen3_vl_30b_a3b_thinking/test_npu_qwen3_vl_30b_a3b_thinking_1p_mmmu.py
|
||||
- name: qwen3_235b_w8a8_8p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/qwen3_235b_a22b/test_npu_qwen3_235b_w8a8_8p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: qwen3_5_397b_w4a8_8p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/qwen3_5_397b/test_npu_qwen3_5_397b_w4a8_8p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
- name: kimi_k2_6_w4a8_8p_in3k5_out1k5_20ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/kimi_k2_6/test_npu_kimi_k2_6_w4a8_8p_in3k5_out1k5_20ms.py
|
||||
test_type: 'perf'
|
||||
# deepseek_v4_flash performance tests
|
||||
- name: deepseek_v4_flash_w8a8_8p_in8k_out1k_50ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/ascend/performance/deepseek_v4_flash/test_npu_deepseek_v4_flash_w8a8_8p_in8k_out1k_50ms.py
|
||||
test_case: test/registered/npu/performance/deepseek_v4_flash/test_npu_deepseek_v4_flash_w8a8_8p_in8k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
uses: ./.github/workflows/nightly-test-npu-e2e-single-node.yml
|
||||
with:
|
||||
@@ -272,26 +314,32 @@ jobs:
|
||||
matrix:
|
||||
test_config:
|
||||
# glm5_1 performance tests
|
||||
- name: glm5_1_w4a8_1p1d_32p_in64k_out1k_50ms
|
||||
prefill_size: 2
|
||||
decode_size: 2
|
||||
router_size: 1
|
||||
test_case: test/registered/npu/performance/glm5_1/test_npu_glm5_1_w4a8_1p1d_32p_in64k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
prefill_decode_deployment: 'separation'
|
||||
- name: glm5_1_w4a8_1p1d_32p_in64k_out1k_50ms_aime26
|
||||
prefill_size: 2
|
||||
decode_size: 2
|
||||
router_size: 1
|
||||
test_case: test/registered/ascend/performance/glm5_1/test_npu_glm5_1_w4a8_1p1d_32p_in64k_out1k_50ms_aime26.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/glm5_1/test_npu_glm5_1_w4a8_1p1d_32p_in64k_out1k_50ms_aime26.py
|
||||
prefill_decode_deployment: 'separation'
|
||||
# mimo_v2_flash performance tests
|
||||
- name: test_npu_mimo_v2_flash_1p1d_12p_in16k_out1_ttft_5s
|
||||
prefill_size: 1
|
||||
decode_size: 1
|
||||
router_size: 1
|
||||
test_case: test/registered/ascend/performance/mimo_v2_flash/test_npu_mimo_v2_flash_1p1d_12p_in16k_out1_ttft_5s.py
|
||||
test_case: test/registered/npu/performance/mimo_v2_flash/test_npu_mimo_v2_flash_1p1d_12p_in16k_out1_ttft_5s.py
|
||||
test_type: 'perf'
|
||||
prefill_decode_deployment: 'separation'
|
||||
- name: test_npu_mimo_v2_flash_1p1d_12p_in16k_out1k_tpot_20ms
|
||||
prefill_size: 1
|
||||
decode_size: 1
|
||||
router_size: 1
|
||||
test_case: test/registered/ascend/performance/mimo_v2_flash/test_npu_mimo_v2_flash_1p1d_12p_in16k_out1k_tpot_20ms.py
|
||||
test_case: test/registered/npu/performance/mimo_v2_flash/test_npu_mimo_v2_flash_1p1d_12p_in16k_out1k_tpot_20ms.py
|
||||
test_type: 'perf'
|
||||
prefill_decode_deployment: 'separation'
|
||||
uses: ./.github/workflows/nightly-test-npu-e2e-multi-node.yml
|
||||
@@ -318,10 +366,13 @@ jobs:
|
||||
matrix:
|
||||
test_config:
|
||||
# kimi_k2_6 performance tests
|
||||
- name: kimi_k2_6_w4a8_16p_in64k_out1k_100ms
|
||||
node_size: 2
|
||||
test_case: test/registered/npu/performance/kimi_k2_6/test_npu_kimi_k2_6_w4a8_16p_in64k_out1k_100ms.py
|
||||
test_type: 'perf'
|
||||
- name: kimi_k2_6_w4a8_16p_in64k_out1k_100ms_aime25
|
||||
node_size: 2
|
||||
test_case: test/registered/ascend/performance/kimi_k2_6/test_npu_kimi_k2_6_w4a8_16p_in64k_out1k_100ms_aime25.py
|
||||
test_type: 'perf'
|
||||
test_case: test/registered/npu/accuracy/kimi_k2_6/test_npu_kimi_k2_6_w4a8_16p_in64k_out1k_100ms_aime25.py
|
||||
uses: ./.github/workflows/nightly-test-npu-e2e-multi-node.yml
|
||||
with:
|
||||
runner: linux-amd64-cpu-8
|
||||
|
||||
@@ -59,7 +59,7 @@ jobs:
|
||||
- "python/sglang/!(multimodal_gen)/**/!(*.md)"
|
||||
- "python/pyproject_npu.toml"
|
||||
- "scripts/ci/npu/npu_ci_install_dependency.sh"
|
||||
- "test/registered/ascend/**"
|
||||
- "test/registered/npu/**"
|
||||
- "test/registered/unit/npu/**"
|
||||
- ".github/workflows/pr-test-npu.yml"
|
||||
multimodal_gen:
|
||||
@@ -461,7 +461,78 @@ jobs:
|
||||
# qwen3_6_27b performance tests
|
||||
- name: qwen3_6_27b_w8a8_1p_in64k_out1k_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in64k_out1k_50ms.py
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in64k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
- name: qwen3_6_27b_1p_in1024x1024_30_out1024_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3_6_27b/test_npu_qwen3_6_27b_1p_in1024x1024_30_out1024_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# qwen3_8b performance tests
|
||||
- name: qwen3_8b_w8a8_1p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3-8b/test_npu_qwen3_8b_w8a8_1p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# qwen3_30b_a3b performance tests
|
||||
- name: qwen3_30b_w8a8_1p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3_30b_a3b/test_npu_qwen3_30b_w8a8_1p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# qwen3_6_35b_a3b performance tests
|
||||
- name: qwen3_6_35b_a3b_1p_in64k_out1k_prefix90_50ms
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/performance/qwen3_6_35b_a3b/test_npu_qwen3_6_35b_a3b_1p_in64k_out1k_prefix90_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# qwen3_vl_8b_thinking accuracy tests
|
||||
- name: qwen3_vl_8b_thinking_1p_mmmu
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/npu/accuracy/qwen3_vl_8b_thinking/test_npu_qwen3_vl_8b_thinking_1p_mmmu.py
|
||||
test_type: 'accuracy'
|
||||
|
||||
# qwen3_32b performance tests
|
||||
- name: qwen3_32b_w8a8_2p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/npu/performance/qwen3_32b/test_npu_qwen3_32b_w8a8_2p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# qwen3_next_80b_a3b performance tests
|
||||
- name: qwen3_next_80b_w8a8_2p_in6k_out1k5_bs16
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/npu/performance/qwen3_next_80b_a3b_instruct/test_npu_qwen3_next_80b_w8a8_2p_in6k_out1k5_bs16.py
|
||||
test_type: 'perf'
|
||||
|
||||
# minimax_m2_5 performance tests
|
||||
- name: minimax_m2_5_w8a8_4p_in64k_out1k_prefix90_50ms
|
||||
runner: linux-aarch64-a3-8
|
||||
test_case: test/registered/npu/performance/minimax_m2_5/test_npu_minimax_m2_5_w8a8_4p_in64k_out1k_prefix90_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# deepseek_v4_flash performance tests
|
||||
- name: deepseek_v4_flash_w8a8_8p_in8k_out1k_50ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/deepseek_v4_flash/test_npu_deepseek_v4_flash_w8a8_8p_in8k_out1k_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# kimi_k2_6 performance tests
|
||||
- name: kimi_k2_6_w4a8_8p_in3k5_out1k5_20ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/kimi_k2_6/test_npu_kimi_k2_6_w4a8_8p_in3k5_out1k5_20ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# qwen3_235b performance tests
|
||||
- name: qwen3_235b_w8a8_8p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/qwen3_235b_a22b/test_npu_qwen3_235b_w8a8_8p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
|
||||
# qwen3_5_397b performance tests
|
||||
- name: qwen3_5_397b_w4a8_8p_in3k5_out1k5_50ms
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/npu/performance/qwen3_5_397b/test_npu_qwen3_5_397b_w4a8_8p_in3k5_out1k5_50ms.py
|
||||
test_type: 'perf'
|
||||
# NPU accuracy tests
|
||||
- name: glm4_7_flash_1p_gsm8k
|
||||
|
||||
Reference in New Issue
Block a user