mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2025-12-09 23:06:10 +08:00
Compare commits
base: xinyun:9420d6768efd7f187d71daa8a5ce1c9bca783cef
xinyun:main
xinyun:wentao-optimize-startup-log-2
xinyun:wentao-fix-test-batch-invariant-failure
xinyun:zhuohan/remove-virtual-engine
xinyun:zhuohan/remove-unnecessary-error-message
xinyun:wentao-fix-mypy-v1
xinyun:woosuk/test-router
xinyun:woosuk/router-nixl
xinyun:moe_dual_stream
xinyun:copilot/disable-batched-triton-kernel
xinyun:zhuohan/moe-kernel-experiment
xinyun:update_from_kv_xfer_finished_race_fix
xinyun:releases/v0.11.1
xinyun:verbose-prime-rl-ci
xinyun:woosuk/remove-req-idx-mapping
xinyun:skip-lmfe-tests
xinyun:releases/v0.11.0
xinyun:releases/v0.10.2
xinyun:codex/remove-vllm-v0-engine-references-from-docs
xinyun:wye-refactor-w8a8-quant
xinyun:debug-logs
xinyun:use-uv-python-for-docker
xinyun:simon-mo-patch-1
xinyun:decode_bench_connector
xinyun:fix_ds_eagle
xinyun:maybe_fix_hang_2
xinyun:fix_hang
xinyun:dbo-cudagraph-size-cherry
xinyun:lwilkinson/cg-support
xinyun:lwilkinson/potential-cutlass-mla-fix
xinyun:amd_dev
xinyun:avoid-double-free
xinyun:woosuk/model-runner-v2
xinyun:support_global_dp_logging
xinyun:khluu/test_latest_feat
xinyun:woosuk/fix-graph-pool
xinyun:codex/remove-raydistributedexecutor-from-v0-engine
xinyun:amd_mori
xinyun:woosuk/minor-worker-fix
xinyun:marlin_gptoss_swiglu
xinyun:split_kv_cache_init
xinyun:copilot/fix-31e676e9-a4af-4ed2-b74d-19d27f0a57b2
xinyun:lwilkinson/eagle-piecewise
xinyun:woosuk/input-prep
xinyun:khluu/nccl
xinyun:copilot/fix-cudagraph-flag-combination
xinyun:debug
xinyun:dependabot/github_actions/actions/checkout-5.0.0
xinyun:il_tool
xinyun:memory-leak-branch
xinyun:khluu/clean_apt
xinyun:woosuk/sampled-token-ids
xinyun:codex/add-auto-max-model-length-setting
xinyun:compile-eplb
xinyun:khluu/use_ccache_premerge
xinyun:woosuk/cleanup-flashinfer
xinyun:khluu/test_fixed_premerge
xinyun:copilot/fix-870996da-9146-438e-9a52-cdc6c1743086
xinyun:copilot/fix-584be906-f283-4e17-8776-c14111357ee7
xinyun:copilot/fix-56244f30-e76a-41ed-beaf-3bc9de22a2c9
xinyun:copilot/fix-c6914add-1b66-46d0-9948-c2e7b6f2259f
xinyun:prune-samplers-test
xinyun:releases/v0.10.1
xinyun:khluu/test_us_east_1
xinyun:lwilkinson/dbo-full-cudagraphs
xinyun:torch-2.8
xinyun:woosuk/flashinfer-swa
xinyun:remove-regression-test
xinyun:remove-metrics-and-tracing-test
xinyun:remove-async-engine-tests
xinyun:fix-v1-test
xinyun:seemethere/cuda_arm64
xinyun:revert-22299-main
xinyun:remove_mamba_ssm
xinyun:woosuk/fa3-swa-cudagraph
xinyun:acc-rate
xinyun:build-flashinfer-aot-wheel
xinyun:releases/v0.10.0
xinyun:wide_ep_working_branch_2
xinyun:wide_ep_working_branch
xinyun:revert-21550-chengji/fix-ci
xinyun:test-debug-lb
xinyun:debug-logging
xinyun:add-nixl-transfer-time-logging
xinyun:tms/distributed_timeout
xinyun:7snzwi-codex/change-default-logging-behavior
xinyun:codex/change-default-logging-behavior
xinyun:nixl-upstreaming
xinyun:benchmark
xinyun:triton-configs
xinyun:fused-moe-tuning-ep
xinyun:mla_decode_any_head
xinyun:gpu_ids2
xinyun:nixl-debug-oh-fixed
xinyun:gpu-ids
xinyun:fix-doc-build
xinyun:dockerfile-nvcc-compress
xinyun:releases/v0.9.2
xinyun:topk_id_hack
xinyun:add-utils
xinyun:minus_x
xinyun:gemma3n-mm
xinyun:deep_full_cudagraph_fix
xinyun:deepep_tweaks
xinyun:fix-precommit
xinyun:releases/v0.9.1
xinyun:mergify/houseroad/config-update
xinyun:lwilkinson/refactor-cmake
xinyun:codex/add-pandas-and-datasets-to-requirements
xinyun:fp8_ep_dp
xinyun:releases/v0.9.0
xinyun:codex/update-arch-overview-md-with-vllm-v1-details
xinyun:benchmark_serving_test
xinyun:pil_image
xinyun:low_latency_opt
xinyun:woosuk-jf
xinyun:disable-sd
xinyun:v0.8.5
xinyun:benchmark-output
xinyun:khluu/test
xinyun:pd_scheduling
xinyun:v0.8.4
xinyun:v1_fix_profiler
xinyun:fix_use_ep
xinyun:dynamo-patch
xinyun:v0.8.3
xinyun:whisper-translate
xinyun:bench-latency
xinyun:khluu/try_moc
xinyun:sampler-env-variable
xinyun:v1-block-table-opt
xinyun:v0.8.2
xinyun:rob-fixes
xinyun:v1-sched-interface-2
xinyun:v0.8.1
xinyun:v0.8.0
xinyun:mamba_tests
xinyun:running-deque
xinyun:bind_kv_caches
xinyun:reduce_scatter_comm
xinyun:amd-ci
xinyun:mla-support-awq-marlin
xinyun:tpu_v1_optimized
xinyun:v0.7.2-staging-branch
xinyun:full_cudagraph
xinyun:mla_cuda_graphs
xinyun:qwen25vl
xinyun:tpu_v1
xinyun:moondream2
xinyun:v1-blocktable-opt
xinyun:correct-docs-cuda-version
xinyun:torch_dynamo
xinyun:optimize-prefix-caching-scheduling
xinyun:fix-hashing-partial-blocks
xinyun:jax-tpu
xinyun:v0.11.1rc1
xinyun:v0.11.0
xinyun:v0.10.2
xinyun:v0.11.0rc6
xinyun:v0.11.0rc5
xinyun:v0.11.0rc4
xinyun:v0.11.0rc3
xinyun:v0.11.0rc2
xinyun:v0.11.1rc0
xinyun:v0.11.0rc1
xinyun:ci/build/22474
xinyun:v0.10.2rc3
xinyun:v0.10.2rc2
xinyun:v0.10.2rc1
xinyun:v0.10.1.1
xinyun:v0.10.1
xinyun:v0.10.1rc1
xinyun:v0.10.0
xinyun:v0.10.0rc2
xinyun:v0.10.0rc1
xinyun:v0.9.2
xinyun:v0.9.2rc2
xinyun:v0.9.2rc1
xinyun:v0.9.1
xinyun:v0.9.1rc2
xinyun:v0.9.1rc1
xinyun:v0.9.0.1
xinyun:v0.9.0
xinyun:v0.8.5.post1
xinyun:v0.8.5
xinyun:v0.8.4
xinyun:v0.8.3
xinyun:v0.8.3rc1
xinyun:v0.8.2
xinyun:v0.8.1
xinyun:v0.8.0
xinyun:v0.8.0rc2
xinyun:v0.8.0rc1
xinyun:v0.7.3
xinyun:v0.7.2
xinyun:v0.7.1
xinyun:v0.7.0
xinyun:v0.6.6.post1
xinyun:v0.6.6
xinyun:v0.6.5
xinyun:v0.6.4.post1
xinyun:v0.6.4
xinyun:v0.6.3.post1
xinyun:v0.6.3
xinyun:v0.6.2
xinyun:v0.6.1.post2
xinyun:v0.6.1.post1
xinyun:v0.6.1
xinyun:v0.6.0
xinyun:v0.5.5
xinyun:v0.5.4
xinyun:v0.5.3.post1
xinyun:v0.5.3
xinyun:v0.5.2
xinyun:v0.5.1
xinyun:v0.5.0.post1
xinyun:v0.5.0
xinyun:v0.4.3
xinyun:v0.4.2
xinyun:v0.4.1
xinyun:v0.4.0.post1
xinyun:v0.4.0
xinyun:v0.3.3
xinyun:v0.3.2
xinyun:v0.3.1
xinyun:v0.3.0
xinyun:v0.2.7
xinyun:v0.2.6
xinyun:v0.2.5
xinyun:v0.2.4
xinyun:v0.2.3
xinyun:v0.2.2
xinyun:v0.2.1.post1
xinyun:v0.2.1
xinyun:v0.2.0
xinyun:v0.1.7
xinyun:v0.1.6
xinyun:v0.1.5
xinyun:v0.1.4
xinyun:v0.1.3
xinyun:v0.1.2
xinyun:v0.1.1
xinyun:v0.1.0
xinyun:submission
...
compare: xinyun:c0c77472cbdd624b4fc8fe2f608f9e6618ccdee2
xinyun:wentao-optimize-startup-log-2
xinyun:wentao-fix-test-batch-invariant-failure
xinyun:main
xinyun:zhuohan/remove-virtual-engine
xinyun:zhuohan/remove-unnecessary-error-message
xinyun:wentao-fix-mypy-v1
xinyun:woosuk/test-router
xinyun:woosuk/router-nixl
xinyun:moe_dual_stream
xinyun:copilot/disable-batched-triton-kernel
xinyun:zhuohan/moe-kernel-experiment
xinyun:update_from_kv_xfer_finished_race_fix
xinyun:releases/v0.11.1
xinyun:verbose-prime-rl-ci
xinyun:woosuk/remove-req-idx-mapping
xinyun:skip-lmfe-tests
xinyun:releases/v0.11.0
xinyun:releases/v0.10.2
xinyun:codex/remove-vllm-v0-engine-references-from-docs
xinyun:wye-refactor-w8a8-quant
xinyun:debug-logs
xinyun:use-uv-python-for-docker
xinyun:simon-mo-patch-1
xinyun:decode_bench_connector
xinyun:fix_ds_eagle
xinyun:maybe_fix_hang_2
xinyun:fix_hang
xinyun:dbo-cudagraph-size-cherry
xinyun:lwilkinson/cg-support
xinyun:lwilkinson/potential-cutlass-mla-fix
xinyun:amd_dev
xinyun:avoid-double-free
xinyun:woosuk/model-runner-v2
xinyun:support_global_dp_logging
xinyun:khluu/test_latest_feat
xinyun:woosuk/fix-graph-pool
xinyun:codex/remove-raydistributedexecutor-from-v0-engine
xinyun:amd_mori
xinyun:woosuk/minor-worker-fix
xinyun:marlin_gptoss_swiglu
xinyun:split_kv_cache_init
xinyun:copilot/fix-31e676e9-a4af-4ed2-b74d-19d27f0a57b2
xinyun:lwilkinson/eagle-piecewise
xinyun:woosuk/input-prep
xinyun:khluu/nccl
xinyun:copilot/fix-cudagraph-flag-combination
xinyun:debug
xinyun:dependabot/github_actions/actions/checkout-5.0.0
xinyun:il_tool
xinyun:memory-leak-branch
xinyun:khluu/clean_apt
xinyun:woosuk/sampled-token-ids
xinyun:codex/add-auto-max-model-length-setting
xinyun:compile-eplb
xinyun:khluu/use_ccache_premerge
xinyun:woosuk/cleanup-flashinfer
xinyun:khluu/test_fixed_premerge
xinyun:copilot/fix-870996da-9146-438e-9a52-cdc6c1743086
xinyun:copilot/fix-584be906-f283-4e17-8776-c14111357ee7
xinyun:copilot/fix-56244f30-e76a-41ed-beaf-3bc9de22a2c9
xinyun:copilot/fix-c6914add-1b66-46d0-9948-c2e7b6f2259f
xinyun:prune-samplers-test
xinyun:releases/v0.10.1
xinyun:khluu/test_us_east_1
xinyun:lwilkinson/dbo-full-cudagraphs
xinyun:torch-2.8
xinyun:woosuk/flashinfer-swa
xinyun:remove-regression-test
xinyun:remove-metrics-and-tracing-test
xinyun:remove-async-engine-tests
xinyun:fix-v1-test
xinyun:seemethere/cuda_arm64
xinyun:revert-22299-main
xinyun:remove_mamba_ssm
xinyun:woosuk/fa3-swa-cudagraph
xinyun:acc-rate
xinyun:build-flashinfer-aot-wheel
xinyun:releases/v0.10.0
xinyun:wide_ep_working_branch_2
xinyun:wide_ep_working_branch
xinyun:revert-21550-chengji/fix-ci
xinyun:test-debug-lb
xinyun:debug-logging
xinyun:add-nixl-transfer-time-logging
xinyun:tms/distributed_timeout
xinyun:7snzwi-codex/change-default-logging-behavior
xinyun:codex/change-default-logging-behavior
xinyun:nixl-upstreaming
xinyun:benchmark
xinyun:triton-configs
xinyun:fused-moe-tuning-ep
xinyun:mla_decode_any_head
xinyun:gpu_ids2
xinyun:nixl-debug-oh-fixed
xinyun:gpu-ids
xinyun:fix-doc-build
xinyun:dockerfile-nvcc-compress
xinyun:releases/v0.9.2
xinyun:topk_id_hack
xinyun:add-utils
xinyun:minus_x
xinyun:gemma3n-mm
xinyun:deep_full_cudagraph_fix
xinyun:deepep_tweaks
xinyun:fix-precommit
xinyun:releases/v0.9.1
xinyun:mergify/houseroad/config-update
xinyun:lwilkinson/refactor-cmake
xinyun:codex/add-pandas-and-datasets-to-requirements
xinyun:fp8_ep_dp
xinyun:releases/v0.9.0
xinyun:codex/update-arch-overview-md-with-vllm-v1-details
xinyun:benchmark_serving_test
xinyun:pil_image
xinyun:low_latency_opt
xinyun:woosuk-jf
xinyun:disable-sd
xinyun:v0.8.5
xinyun:benchmark-output
xinyun:khluu/test
xinyun:pd_scheduling
xinyun:v0.8.4
xinyun:v1_fix_profiler
xinyun:fix_use_ep
xinyun:dynamo-patch
xinyun:v0.8.3
xinyun:whisper-translate
xinyun:bench-latency
xinyun:khluu/try_moc
xinyun:sampler-env-variable
xinyun:v1-block-table-opt
xinyun:v0.8.2
xinyun:rob-fixes
xinyun:v1-sched-interface-2
xinyun:v0.8.1
xinyun:v0.8.0
xinyun:mamba_tests
xinyun:running-deque
xinyun:bind_kv_caches
xinyun:reduce_scatter_comm
xinyun:amd-ci
xinyun:mla-support-awq-marlin
xinyun:tpu_v1_optimized
xinyun:v0.7.2-staging-branch
xinyun:full_cudagraph
xinyun:mla_cuda_graphs
xinyun:qwen25vl
xinyun:tpu_v1
xinyun:moondream2
xinyun:v1-blocktable-opt
xinyun:correct-docs-cuda-version
xinyun:torch_dynamo
xinyun:optimize-prefix-caching-scheduling
xinyun:fix-hashing-partial-blocks
xinyun:jax-tpu
xinyun:v0.11.1rc1
xinyun:v0.11.0
xinyun:v0.10.2
xinyun:v0.11.0rc6
xinyun:v0.11.0rc5
xinyun:v0.11.0rc4
xinyun:v0.11.0rc3
xinyun:v0.11.0rc2
xinyun:v0.11.1rc0
xinyun:v0.11.0rc1
xinyun:ci/build/22474
xinyun:v0.10.2rc3
xinyun:v0.10.2rc2
xinyun:v0.10.2rc1
xinyun:v0.10.1.1
xinyun:v0.10.1
xinyun:v0.10.1rc1
xinyun:v0.10.0
xinyun:v0.10.0rc2
xinyun:v0.10.0rc1
xinyun:v0.9.2
xinyun:v0.9.2rc2
xinyun:v0.9.2rc1
xinyun:v0.9.1
xinyun:v0.9.1rc2
xinyun:v0.9.1rc1
xinyun:v0.9.0.1
xinyun:v0.9.0
xinyun:v0.8.5.post1
xinyun:v0.8.5
xinyun:v0.8.4
xinyun:v0.8.3
xinyun:v0.8.3rc1
xinyun:v0.8.2
xinyun:v0.8.1
xinyun:v0.8.0
xinyun:v0.8.0rc2
xinyun:v0.8.0rc1
xinyun:v0.7.3
xinyun:v0.7.2
xinyun:v0.7.1
xinyun:v0.7.0
xinyun:v0.6.6.post1
xinyun:v0.6.6
xinyun:v0.6.5
xinyun:v0.6.4.post1
xinyun:v0.6.4
xinyun:v0.6.3.post1
xinyun:v0.6.3
xinyun:v0.6.2
xinyun:v0.6.1.post2
xinyun:v0.6.1.post1
xinyun:v0.6.1
xinyun:v0.6.0
xinyun:v0.5.5
xinyun:v0.5.4
xinyun:v0.5.3.post1
xinyun:v0.5.3
xinyun:v0.5.2
xinyun:v0.5.1
xinyun:v0.5.0.post1
xinyun:v0.5.0
xinyun:v0.4.3
xinyun:v0.4.2
xinyun:v0.4.1
xinyun:v0.4.0.post1
xinyun:v0.4.0
xinyun:v0.3.3
xinyun:v0.3.2
xinyun:v0.3.1
xinyun:v0.3.0
xinyun:v0.2.7
xinyun:v0.2.6
xinyun:v0.2.5
xinyun:v0.2.4
xinyun:v0.2.3
xinyun:v0.2.2
xinyun:v0.2.1.post1
xinyun:v0.2.1
xinyun:v0.2.0
xinyun:v0.1.7
xinyun:v0.1.6
xinyun:v0.1.5
xinyun:v0.1.4
xinyun:v0.1.3
xinyun:v0.1.2
xinyun:v0.1.1
xinyun:v0.1.0
xinyun:submission
3 Commits
9420d6768e
...
c0c77472cb
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
|
|
c0c77472cb
|
Merge 6ab025c10abed1dc5427fd49c34f6a45cb6c6235 into 6c9fdbf7258146a9e335c50aab12969cd95e9227 | ||
|
|
6ab025c10a
|
Merge branch 'main' into fix_ut | ||
|
|
0700fdddc7 |
use FP1 instead
Signed-off-by: Fanli Lin <fanli.lin@intel.com> |
1 changed files with 1 additions and 1 deletions
|
|
@ -11,7 +11,7 @@ PROMPT = "Hello my name is Robert and I"
|
|||
|
||||
@pytest.fixture(scope="module")
|
||||
def llm() -> LLM:
|
||||
return LLM(MODEL, enforce_eager=True)
|
||||
return LLM(MODEL, enforce_eager=True, dtype="half")
|
||||
|
||||
|
||||
def test_n_gt_1(llm):
|
||||
|
|
|
|||
Write
Preview
Loading…
x
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.