Skip to content

Actions: vectorch-ai/ScaleLLM

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
26 workflow run results
26 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

added boost dependency to fix build error.
Build and test #30: Commit 586856d pushed by guocuimi
November 22, 2023 18:27 15m 13s main
November 22, 2023 18:27 15m 13s
replaced libevhtp with boost asio for http server to avoid epoll_wait…
Build and test #29: Commit 66e016d pushed by guocuimi
November 22, 2023 08:42 10m 32s main
November 22, 2023 08:42 10m 32s
use temperature_penalty kernel for temperature logits processor
Build and test #28: Commit 3ddaa44 pushed by guocuimi
November 22, 2023 01:44 7m 28s main
November 22, 2023 01:44 7m 28s
added args overrider that allows override any model args with command…
Build and test #27: Commit 7e89647 pushed by guocuimi
November 22, 2023 00:06 15m 6s main
November 22, 2023 00:06 15m 6s
set up QEMU and Buildx for multiple platforms
Publish docker image #15: Commit caa8ab2 pushed by guocuimi
November 15, 2023 07:29 31m 46s v0.0.2
November 15, 2023 07:29 31m 46s
added 'disable_custom_kernels' gflags to allow disable all custom ker…
Build and test #26: Commit 149b943 pushed by guocuimi
November 11, 2023 23:32 7m 56s main
November 11, 2023 23:32 7m 56s
fix: always use float32 for cpu.
Build and test #25: Commit bb73a8c pushed by guocuimi
November 9, 2023 23:14 7m 10s main
November 9, 2023 23:14 7m 10s
misc: added chat api column in supported models and only build scalel…
Build and test #23: Commit e5c53ff pushed by guocuimi
November 9, 2023 16:59 9m 1s main
November 9, 2023 16:59 9m 1s
return 503 if server is not running or stopping for health endpoint.
Build and test #22: Commit a6b51e1 pushed by guocuimi
November 9, 2023 07:56 7m 24s main
November 9, 2023 07:56 7m 24s
added chat templates for aquila, internlm and mistral.
Build and test #21: Commit ccf8391 pushed by guocuimi
November 9, 2023 06:21 7m 2s main
November 9, 2023 06:21 7m 2s
upgrade vllm attention kernel to use paged attention v2 for long sequ…
Build and test #20: Commit 5618234 pushed by guocuimi
November 9, 2023 04:50 7m 18s main
November 9, 2023 04:50 7m 18s
remove unused variables
Build and test #19: Commit c0fd500 pushed by guocuimi
November 9, 2023 00:49 7m 21s main
November 9, 2023 00:49 7m 21s
use current cuda stream for all kernels.
Build and test #18: Commit 95f908a pushed by guocuimi
November 9, 2023 00:19 7m 18s main
November 9, 2023 00:19 7m 18s
added Yi into supported model list and disabled flaky unittests
Build and test #17: Commit 908b3ab pushed by guocuimi
November 8, 2023 01:38 9m 52s main
November 8, 2023 01:38 9m 52s
fixed top_k tensor type and added unittests.
Build and test #16: Commit c1acd85 pushed by guocuimi
November 8, 2023 01:24 16m 28s main
November 8, 2023 01:24 16m 28s
Merge pull request #14 from vectorch-ai/v0.0.2
Build and test #14: Commit 706f9e1 pushed by guocuimi
November 7, 2023 19:34 7m 25s main
November 7, 2023 19:34 7m 25s
handle group_size == -1 for gptq quantized weights. tested with 'TheB…
Build and test #12: Commit 95b1f4a pushed by guocuimi
November 7, 2023 06:36 7m 30s main
November 7, 2023 06:36 7m 30s
only prepend bos token for llama2.
Build and test #11: Commit 98afb3e pushed by guocuimi
November 7, 2023 05:55 7m 21s main
November 7, 2023 05:55 7m 21s
added Yi model support. tested with '01-ai/Yi-6B'
Build and test #10: Commit 684d410 pushed by guocuimi
November 7, 2023 05:28 7m 33s main
November 7, 2023 05:28 7m 33s
updated README.md for more details.
Build and publish docker image to Docker Hub #16: Commit c784b0d pushed by guocuimi
November 6, 2023 23:21 30m 59s v0.0.1
November 6, 2023 23:21 30m 59s
updated README.md for more details.
Build and test #9: Commit c784b0d pushed by guocuimi
November 6, 2023 23:14 7m 32s main
November 6, 2023 23:14 7m 32s
misc: added 'auto' option for device, fixed the build type for cargo …
Build and test #8: Commit 266c01a pushed by guocuimi
November 5, 2023 05:40 7m 40s main
November 5, 2023 05:40 7m 40s
added Dockerfile for development and workflows for CI and docker.
Build and test #7: Commit 7d83030 pushed by guocuimi
November 4, 2023 06:44 45m 47s main
November 4, 2023 06:44 45m 47s
add release tag for docker image
Build and test #6: Commit 840e103 pushed by guocuimi
November 4, 2023 04:59 35m 46s main
November 4, 2023 04:59 35m 46s
update tags format.
Build and test #4: Commit 5424d64 pushed by guocuimi
November 4, 2023 04:40 19m 49s main
November 4, 2023 04:40 19m 49s