Commit graph

  • 44c05c6e7d add vision instruct models for fireworks Dalton Flanagan 2024-10-27 17:54:54 +00:00
  • 91e5ad18b0 remove prints, cleanup braintrust in this branch Xi Yan 2024-10-25 17:39:05 -07:00
  • 9b410a87bf extract score regex to llm context Xi Yan 2024-10-25 17:03:01 -07:00
  • 247a53d393 test full generation + eval Xi Yan 2024-10-25 16:52:59 -07:00
  • 16620a8185 llm as judge, move folders Xi Yan 2024-10-25 16:41:36 -07:00
  • 9b85d9a841
    completion() for fireworks (#329) Dinesh Yeduguru 2024-10-25 16:12:10 -07:00
  • d882d46618 completion() for fireworks Dinesh Yeduguru 2024-10-25 15:41:53 -07:00
  • bf8bc7a781 wip scoring refactor Xi Yan 2024-10-25 15:03:03 -07:00
  • 7ec79f3b9d
    completion() for together (#324) Dinesh Yeduguru 2024-10-25 14:21:12 -07:00
  • 8bbe13e8d1 fix client building Dinesh Yeduguru 2024-10-25 11:28:22 -07:00
  • 8daf2e78be test fixes Dinesh Yeduguru 2024-10-25 10:46:28 -07:00
  • d83715be9b completion() for together Dinesh Yeduguru 2024-10-25 10:38:04 -07:00
  • 8a74e400d6
    Update getting_started.md Xi Yan 2024-10-25 13:30:33 -07:00
  • f168752bba
    Update getting_started.md Xi Yan 2024-10-25 13:27:43 -07:00
  • abdf7cddf3
    [Evals API][4/n] evals with generation meta-reference impl (#303) Xi Yan 2024-10-25 13:12:39 -07:00
  • 443ea8872a sample eval run.yaml Xi Yan 2024-10-25 13:10:56 -07:00
  • 426d821e7f Bump version to 0.0.46 Ashwin Bharambe 2024-10-25 13:10:55 -07:00
  • 204a5c6d91 remove custom type Xi Yan 2024-10-25 13:02:56 -07:00
  • 4e2870445b
    [Evals API][5/n] fixes to generate openapi spec (#323) Xi Yan 2024-10-25 13:00:40 -07:00
  • 81ebd1ea92 typing comment, dataset -> dataset_id Xi Yan 2024-10-25 12:59:47 -07:00
  • c05fbf14b3
    Added hadamard transform for spinquant (#326) Sachin Mehta 2024-10-25 12:58:48 -07:00
  • cd39509e56 pre-commit Ashwin Bharambe 2024-10-25 12:58:17 -07:00
  • 575e51eb76 Merge branch 'evals_6' into evals_7 Xi Yan 2024-10-25 12:55:51 -07:00
  • d95bef7f2e Merge branch 'main' into evals_6 Xi Yan 2024-10-25 12:55:28 -07:00
  • d0b6894a41 Use enum.value to check against str Ashwin Bharambe 2024-10-25 12:53:32 -07:00
  • 07f9bf723f
    fix broken --list-templates with adding build.yaml files for packaging (#327) Xi Yan 2024-10-25 12:51:22 -07:00
  • 6100b02ff5 fix readmes Xi Yan 2024-10-25 12:49:33 -07:00
  • 474101a9f7 fix build from templates CLI Xi Yan 2024-10-25 12:44:04 -07:00
  • 7e60a9ca6d remove image_type in templates Xi Yan 2024-10-25 12:37:15 -07:00
  • 8e0a4e2885 Added an assertion for model args Sachin Mehta 2024-10-25 12:33:27 -07:00
  • db4f18099f Changed from config to model_args Sachin Mehta 2024-10-25 12:24:50 -07:00
  • 65d914d7b9 change everything to docker build.yaml Xi Yan 2024-10-25 12:19:17 -07:00
  • 98ba12eb15 precommit Xi Yan 2024-10-25 12:15:21 -07:00
  • 37010c4b58 symlink Xi Yan 2024-10-25 12:12:48 -07:00
  • d17c2e5675 symlink Xi Yan 2024-10-25 12:12:13 -07:00
  • 02b2e1c4a4 manifest Xi Yan 2024-10-25 12:10:15 -07:00
  • 32df363b48 fix templates Xi Yan 2024-10-25 12:09:45 -07:00
  • 19adb4070a add build files to templates Xi Yan 2024-10-25 12:08:32 -07:00
  • 94f4f0331d move build.yaml to templates, symlink in distributions Xi Yan 2024-10-25 11:54:09 -07:00
  • 753e8790a3 Update docker build flow a little Ashwin Bharambe 2024-10-25 10:06:21 -07:00
  • 3c3c6e11ad Move function around Ashwin Bharambe 2024-10-25 09:17:33 -07:00
  • 392c23e557 Update docker_base for meta-reference-gpu Ashwin Bharambe 2024-10-25 09:13:33 -07:00
  • be3adb0964 Make vllm inference better Ashwin Bharambe 2024-10-24 22:30:49 -07:00
  • b4a5176a54 update MANIFEST Xi Yan 2024-10-25 11:54:51 -07:00
  • 56f9b7d5d6 move build.yaml to templates, symlink in distributions Xi Yan 2024-10-25 11:54:09 -07:00
  • 93472042f8 Added hadamard transform for spinquant Sachin Mehta 2024-10-25 11:48:24 -07:00
  • 52fe165db8 address comments Xi Yan 2024-10-25 11:08:40 -07:00
  • e83ce36780 fix precommit hook ci failure Dinesh Yeduguru 2024-10-25 11:03:05 -07:00
  • 81ed0327f3 tmp wip Xi Yan 2024-10-25 10:43:22 -07:00
  • b543785c92
    Merge branch 'meta-llama:main' into merge_conflict karthikgutha 2024-10-25 10:23:10 -07:00
  • afae4e3d8e Update docker build flow a little Ashwin Bharambe 2024-10-25 10:06:21 -07:00
  • 5bed6c276c Move function around Ashwin Bharambe 2024-10-25 09:17:33 -07:00
  • a387ca22e2 Update docker_base for meta-reference-gpu Ashwin Bharambe 2024-10-25 09:13:33 -07:00
  • 14cd065b6c
    Merge branch 'main' into merge_conflict karthikgutha 2024-10-25 07:45:36 -07:00
  • 58c3c45f19 workaround list templates command Xi Yan 2024-10-24 23:05:26 -07:00
  • 70d59b0f5d Make vllm inference better Ashwin Bharambe 2024-10-24 22:30:49 -07:00
  • cb43caa2c3 start_container.sh prefix llamastack->distribution name Xi Yan 2024-10-24 21:29:07 -07:00
  • 6b0baa6d53 scoring_function -> scoring_fn rename, scorer -> scoring_fn rename Xi Yan 2024-10-24 21:06:38 -07:00
  • df141b6ef3
    Fix for get_agents_session (#300) Sarthak Deshpande 2024-10-25 07:06:27 +05:30
  • ec7c8f95de generate openapi Xi Yan 2024-10-24 17:41:15 -07:00
  • cdfd584a8f Merge branch 'main' into evals_6 Xi Yan 2024-10-24 17:29:22 -07:00
  • b6d8246b82
    added templates and enhanced readme (#307) Justin Lee 2024-10-24 17:07:06 -07:00
  • 2372de70f8 error checking Xi Yan 2024-10-24 16:16:34 -07:00
  • f6340a47d1 inclusion->subsetof Xi Yan 2024-10-24 16:13:49 -07:00
  • 29e48cc5c1 chatcompletion & completion input type validation Xi Yan 2024-10-24 16:11:25 -07:00
  • 6c49846ecc added templates and enhanced readme Justin Lee 2024-10-24 16:05:33 -07:00
  • 3e1c3fdb3f
    completion() for tgi (#295) Dinesh Yeduguru 2024-10-24 16:02:41 -07:00
  • e468e23249
    Merge branch 'main' into evals_6 Xi Yan 2024-10-24 14:59:41 -07:00
  • 997e3003b9 Update getting_started.md Xi Yan 2024-10-24 14:19:35 -07:00
  • dc6393c271 update manifest for build templates Xi Yan 2024-10-24 14:04:13 -07:00
  • 2bea4bf801 Handle both ipv6 and ipv4 interfaces together Ashwin Bharambe 2024-10-24 13:36:41 -07:00
  • 72755da634 Bump version to 0.0.45 Ashwin Bharambe 2024-10-24 12:14:18 -07:00
  • 7b1a45ee0f Fix score threshold in faiss Ashwin Bharambe 2024-10-24 12:11:58 -07:00
  • 86f1efa680 Small updates to quantization config Ashwin Bharambe 2024-10-24 12:08:43 -07:00
  • 1721d91c95 Update iOS inference instructions for new quantization Dalton Flanagan 2024-10-24 14:47:27 -04:00
  • cb84034567
    [Evals API][3/n] scoring_functions / scoring meta-reference implementations (#296) Xi Yan 2024-10-24 14:52:30 -07:00
  • d4887fc746 address comments Xi Yan 2024-10-24 14:49:02 -07:00
  • f7658f9c3a fix assert Dinesh Yeduguru 2024-10-24 14:46:21 -07:00
  • 9bf1388429 actually test strutured output in completion Dinesh Yeduguru 2024-10-24 14:44:31 -07:00
  • e70420a06e
    Update getting_started.md Xi Yan 2024-10-24 14:19:35 -07:00
  • 8615bc9e08 update manifest for build templates Xi Yan 2024-10-24 14:04:13 -07:00
  • ba0186f2c8 refactor Xi Yan 2024-10-24 14:00:41 -07:00
  • 94728d6983 Handle both ipv6 and ipv4 interfaces together Ashwin Bharambe 2024-10-24 13:36:41 -07:00
  • 3db1b3fbcd rebase name Xi Yan 2024-10-24 13:53:41 -07:00
  • 97ca72288c Merge branch 'evals_5' into evals_6 Xi Yan 2024-10-24 13:53:00 -07:00
  • 6053b8dd34 scoring function def rename Xi Yan 2024-10-24 13:51:11 -07:00
  • 689990b48b Merge branch 'evals_5' into evals_6 Xi Yan 2024-10-24 13:06:11 -07:00
  • 42bac85e1f bugfix Xi Yan 2024-10-24 12:16:28 -07:00
  • 0538cc297e Bump version to 0.0.45 Ashwin Bharambe 2024-10-24 12:14:18 -07:00
  • 205bcfdd4e Fix score threshold in faiss Ashwin Bharambe 2024-10-24 12:11:58 -07:00
  • 24dce9cb7a minor typing Xi Yan 2024-10-24 12:08:57 -07:00
  • 161aef0aae Small updates to quantization config Ashwin Bharambe 2024-10-24 12:08:43 -07:00
  • 32a496ab0f Merge branch 'evals_5' into evals_6 Xi Yan 2024-10-24 12:01:41 -07:00
  • a3a8f32541 add all rows scores to ScoringResult Xi Yan 2024-10-24 11:53:15 -07:00
  • 8eceebec98
    Update iOS inference instructions for new quantization Dalton Flanagan 2024-10-24 14:47:27 -04:00
  • 737fcb795f evals with generation Xi Yan 2024-10-24 11:30:13 -07:00
  • 071dba8871 Merge branch 'main' into evals_5 Xi Yan 2024-10-24 09:18:15 -07:00
  • 8aa8847b4a Bump version to 0.0.44 Ashwin Bharambe 2024-10-24 08:41:39 -07:00
  • 7afe51c84d
    New quantized models (#301) Ashwin Bharambe 2024-10-24 08:38:56 -07:00
  • 335c2561fa New quantized models Ashwin Bharambe 2024-10-24 08:37:26 -07:00