{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":701226931,"defaultBranch":"main","name":"optimum-nvidia","ownerLogin":"huggingface","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2023-10-06T07:40:55.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/25720743?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1720356915.0","currentOid":""},"activityList":{"items":[{"before":"37f7e393bb2d8243d11d5a48fd805f05aec95636","after":"b421cc6acc5f8f1f635cf5e9ef41e4755b2a9341","ref":"refs/heads/ci-l4","pushedAt":"2024-07-07T22:23:37.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Allow restoring from local dir","shortMessageHtmlLink":"Allow restoring from local dir"}},{"before":"5ac3547c7f4e647b0d9a1b2929d1947a1df067a8","after":"37f7e393bb2d8243d11d5a48fd805f05aec95636","ref":"refs/heads/ci-l4","pushedAt":"2024-07-07T21:37:28.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Remove quantization for now","shortMessageHtmlLink":"Remove quantization for now"}},{"before":"9e4f0a8fa30678d0d9b02da9eaff0aed44b4dd2b","after":"5ac3547c7f4e647b0d9a1b2929d1947a1df067a8","ref":"refs/heads/ci-l4","pushedAt":"2024-07-07T21:29:05.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Ensure torch dependency constraints match TRTLLM","shortMessageHtmlLink":"Ensure torch dependency constraints match TRTLLM"}},{"before":"077b93c8bf061ef228b7fba3749fee14f4790909","after":"9e4f0a8fa30678d0d9b02da9eaff0aed44b4dd2b","ref":"refs/heads/ci-l4","pushedAt":"2024-07-07T21:15:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Pin TensorRT-LLM to first 0.12 beta version","shortMessageHtmlLink":"Pin TensorRT-LLM to first 0.12 beta version"}},{"before":"ffdf7f0982d6f3bce6ce094868cb534eb09bc78e","after":"077b93c8bf061ef228b7fba3749fee14f4790909","ref":"refs/heads/ci-l4","pushedAt":"2024-07-07T21:11:55.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Pin TensorRT-LLM version","shortMessageHtmlLink":"Pin TensorRT-LLM version"}},{"before":"89c4b8851ae3e2235d2e357c0d48947d18b70ccc","after":"ffdf7f0982d6f3bce6ce094868cb534eb09bc78e","ref":"refs/heads/ci-l4","pushedAt":"2024-07-07T21:00:35.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"remove third-party trtllm","shortMessageHtmlLink":"remove third-party trtllm"}},{"before":"721cb657cb147720c52483c34d24a3849b84749c","after":null,"ref":"refs/heads/ci-new-l4-runners","pushedAt":"2024-07-07T12:55:15.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"}},{"before":null,"after":"89c4b8851ae3e2235d2e357c0d48947d18b70ccc","ref":"refs/heads/ci-l4","pushedAt":"2024-07-07T12:55:07.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Update CI to use new workflow and silicon.","shortMessageHtmlLink":"Update CI to use new workflow and silicon."}},{"before":"80cad50908d496916bf0a2391c5829fe994d3fb5","after":null,"ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-07-07T12:50:05.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"}},{"before":"714734fe05fd4fbf74bdf19d1676a2cd0d7b8eb8","after":"011b5a9dada1bd314e1c9de3b5b6455a29dcba42","ref":"refs/heads/main","pushedAt":"2024-07-07T12:50:02.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Refactor the overall Hugging Face -> TRTLLM export workflow (#133)\n\n* Initial mixtral support\r\n\r\n* feat(mixtral): map the correct num_local_experts config key for the MOE config\r\n\r\n* feat(mixtral): allow to specify TP/PP configurations when allocating the model\r\n\r\n* feat(mixtral): Expose tp/pp in examples/cli\r\n\r\n* feat(mixtral): Remove config attributes from the model_kwargs to avoid setting many dupplicates\r\n\r\n* feat(hub): always do weight layout conversion on CPU memory\r\n\r\n* feat(mixtral): enable MOE config conversion from transformers\r\n\r\n* feat(parallelism): Enable providing TP/PP/MOE parallelism args\r\n\r\n* feat(parallelism): Enable forwarding tp/pp args to trtllm-build\r\n\r\n* feat(converter): Introduce base for TRTModelConverter\r\n\r\n* Upgrade huggingface-hub dependency to 0.23.0\r\n\r\n* feat(hub): Initial refactoring for clear separation of concerns\r\n\r\n* feat(hub): Rework the overall separation of concern for the hub and exporting\r\n\r\n* feat(hub): Working for all non-Whisper model\r\n\r\n* feat(hub): Disable whisper for now\r\n\r\n* feat(trtllm) : Update trtllm to 0.10.0\r\n\r\n* feat(deps) : Ping hf-transfer to 0.1.6\r\n\r\n* feat(quant): Rework overall quantization schema\r\n\r\n* feat(misc): Failed name refactoring leaving untouched imports ...\r\n\r\n* feat(hub): Expose device_map=\"auto\"\r\n\r\n* feat(chore): quality\r\n\r\n* feat(hub): expose device_map to enable auto-parallel\r\n\r\n* feat(docker): Use repo variable for image namelocal\r\n\r\n* feat(build): Validate new workflow for building engines\r\n\r\n* feat(deps): Move to TRTLLM 0.11 preversion\r\n\r\n* feat(deps): Use the new executor api for running LLMs\r\n\r\n* feat(kvcache): Use floor when computing the number of tokens to store in the kvcache\r\n\r\n* feat(ifb): Enable async generation with in-flight batching support\r\n\r\n* feat(misc): Add better typing return info for AutoModelCausalLM\r\n\r\n* feat(chore): quality\r\n\r\n* feat(misc): Remove padding reference in examples\r\n\r\n* Ensure building all the ranks in disitributed settisgs\r\n\r\n* Update hub tests","shortMessageHtmlLink":"Refactor the overall Hugging Face -&gt; TRTLLM export workflow (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2339229860\" data-permission-text=\"Title is private\" data-url=\"https://github.com/huggingface/optimum-nvidia/issues/133\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/huggingface/optimum-nvidia/pull/133/hovercard\" href=\"https://github.com/huggingface/optimum-nvidia/pull/133\">#133</a>)"}},{"before":"474fcced53fe4c3fb97786804aa232ded4f8ae31","after":"80cad50908d496916bf0a2391c5829fe994d3fb5","ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-07-07T12:31:15.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Update hub tests","shortMessageHtmlLink":"Update hub tests"}},{"before":"6ae816e079262a9aebab99db660d91bf05eb22c7","after":"474fcced53fe4c3fb97786804aa232ded4f8ae31","ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-07-05T11:38:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Ensure building all the ranks in disitributed settisgs","shortMessageHtmlLink":"Ensure building all the ranks in disitributed settisgs"}},{"before":"abbf15491472367db247302ae99b7e0da2553c9d","after":"721cb657cb147720c52483c34d24a3849b84749c","ref":"refs/heads/ci-new-l4-runners","pushedAt":"2024-06-25T14:26:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(trtllm): Update pytorch to at least 2.3.0","shortMessageHtmlLink":"feat(trtllm): Update pytorch to at least 2.3.0"}},{"before":"f730b61626dcfc32b665ff2c85a2fe78275ee183","after":"abbf15491472367db247302ae99b7e0da2553c9d","ref":"refs/heads/ci-new-l4-runners","pushedAt":"2024-06-25T14:23:29.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(trtllm): Force TRTLLM to 0.11.0.dev2024062500","shortMessageHtmlLink":"feat(trtllm): Force TRTLLM to 0.11.0.dev2024062500"}},{"before":"dbe261a6b482352dc0564ff5ef9fd6bd653a1ec2","after":"f730b61626dcfc32b665ff2c85a2fe78275ee183","ref":"refs/heads/ci-new-l4-runners","pushedAt":"2024-06-25T13:43:23.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(tests): Update CUDA image base version to 12.4.1","shortMessageHtmlLink":"feat(tests): Update CUDA image base version to 12.4.1"}},{"before":null,"after":"dbe261a6b482352dc0564ff5ef9fd6bd653a1ec2","ref":"refs/heads/ci-new-l4-runners","pushedAt":"2024-06-25T13:31:52.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(tests): Move to k8s cluster for CI/CD","shortMessageHtmlLink":"feat(tests): Move to k8s cluster for CI/CD"}},{"before":"8f27b995633f507242b7e8f30e39b12736bc2488","after":"6ae816e079262a9aebab99db660d91bf05eb22c7","ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-06-25T10:01:07.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(misc): Remove padding reference in examples","shortMessageHtmlLink":"feat(misc): Remove padding reference in examples"}},{"before":"c9c8f984a9412c0a2afbe21523f1c6ed86999561","after":"8f27b995633f507242b7e8f30e39b12736bc2488","ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-06-25T09:58:25.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(chore): quality","shortMessageHtmlLink":"feat(chore): quality"}},{"before":"7c76a4d4fde3cd9662ef60375a1d579b5c890748","after":"c9c8f984a9412c0a2afbe21523f1c6ed86999561","ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-06-25T09:55:26.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(misc): Add better typing return info for AutoModelCausalLM","shortMessageHtmlLink":"feat(misc): Add better typing return info for AutoModelCausalLM"}},{"before":"a8a5c7d5a871e3f37e18fa807bf36c824893039a","after":"7c76a4d4fde3cd9662ef60375a1d579b5c890748","ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-06-25T09:49:04.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(ifb): Enable async generation with in-flight batching support","shortMessageHtmlLink":"feat(ifb): Enable async generation with in-flight batching support"}},{"before":"54e361d226a387935b5da4b5ffba4e3472ad3968","after":"a8a5c7d5a871e3f37e18fa807bf36c824893039a","ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-06-24T12:42:40.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(deps): Use the new executor api for running LLMs","shortMessageHtmlLink":"feat(deps): Use the new executor api for running LLMs"}},{"before":"210fab0455f94b17f793503751730729c55426eb","after":null,"ref":"refs/heads/ci_docker_registry_push","pushedAt":"2024-06-24T07:46:39.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"}},{"before":"d03088b6e3df94c1b18b51b2d1b3f3c822d7fca9","after":"714734fe05fd4fbf74bdf19d1676a2cd0d7b8eb8","ref":"refs/heads/main","pushedAt":"2024-06-24T07:46:36.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"Enable automatic build of container at each release (#137)\n\n* feat(docker): enable automatic build of container at each release\r\n\r\n* feat(docker): Use repo variable for image name","shortMessageHtmlLink":"Enable automatic build of container at each release (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2353362990\" data-permission-text=\"Title is private\" data-url=\"https://github.com/huggingface/optimum-nvidia/issues/137\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/huggingface/optimum-nvidia/pull/137/hovercard\" href=\"https://github.com/huggingface/optimum-nvidia/pull/137\">#137</a>)"}},{"before":"626cce338a2068f424342f651afaafc54a48d877","after":"210fab0455f94b17f793503751730729c55426eb","ref":"refs/heads/ci_docker_registry_push","pushedAt":"2024-06-14T13:17:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(docker): Use repo variable for image name","shortMessageHtmlLink":"feat(docker): Use repo variable for image name"}},{"before":null,"after":"626cce338a2068f424342f651afaafc54a48d877","ref":"refs/heads/ci_docker_registry_push","pushedAt":"2024-06-14T13:15:33.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(docker): enable automatic build of container at each release","shortMessageHtmlLink":"feat(docker): enable automatic build of container at each release"}},{"before":"28237467b5a21d76196bc5137358ed2a115189ba","after":null,"ref":"refs/heads/trufflehog","pushedAt":"2024-06-14T08:48:47.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"}},{"before":"148415f24f9b8ade7f483d9243cb1cfdc1290e40","after":"d03088b6e3df94c1b18b51b2d1b3f3c822d7fca9","ref":"refs/heads/main","pushedAt":"2024-06-14T08:48:45.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(security): enable trufflehog scanner CI on GA (#136)","shortMessageHtmlLink":"feat(security): enable trufflehog scanner CI on GA (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2352795157\" data-permission-text=\"Title is private\" data-url=\"https://github.com/huggingface/optimum-nvidia/issues/136\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/huggingface/optimum-nvidia/pull/136/hovercard\" href=\"https://github.com/huggingface/optimum-nvidia/pull/136\">#136</a>)"}},{"before":null,"after":"28237467b5a21d76196bc5137358ed2a115189ba","ref":"refs/heads/trufflehog","pushedAt":"2024-06-14T08:02:46.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(security): enable trufflehog scanner CI on GA","shortMessageHtmlLink":"feat(security): enable trufflehog scanner CI on GA"}},{"before":"e1c601bbedc89dbf75d1a558764489c22c3e9036","after":"54e361d226a387935b5da4b5ffba4e3472ad3968","ref":"refs/heads/refactor-trtllm-workflow","pushedAt":"2024-06-14T07:41:26.000Z","pushType":"push","commitsCount":8,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(hub): expose device_map to enable auto-parallel","shortMessageHtmlLink":"feat(hub): expose device_map to enable auto-parallel"}},{"before":"50bbb1635fa7ad875ee570a48217832ea081252d","after":"00da88a7a121adf3a47f88cd2bb8cdf226349c30","ref":"refs/heads/upgrade_trtllm_0100","pushedAt":"2024-06-11T15:46:35.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"mfuntowicz","name":"Funtowicz Morgan","path":"/mfuntowicz","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/2241520?s=80&v=4"},"commit":{"message":"feat(quality)","shortMessageHtmlLink":"feat(quality)"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEeQv2xAA","startCursor":null,"endCursor":null}},"title":"Activity · huggingface/optimum-nvidia"}