Commit Graph

  • e3ded361b2
    feat(ci): improve CI speed (#94) OlivierDehaene 2023-03-03 15:07:27 +0100
  • 18e8f03070 faster builds OlivierDehaene 2023-03-03 14:31:27 +0100
  • 787e20a199 should work OlivierDehaene 2023-03-03 12:57:16 +0100
  • dfb8a6f09c fix OlivierDehaene 2023-03-03 11:46:32 +0100
  • a375ab2976 feat(ci): add sccache OlivierDehaene 2023-03-03 11:42:57 +0100
  • 2d39f199ae
    feat(server): update to hf_transfer==0.1.2 (#93) OlivierDehaene 2023-03-03 11:26:27 +0100
  • 1f4390a929 feat(server): update to hf_transfer==0.1.2 OlivierDehaene 2023-03-03 10:58:13 +0100
  • 9b8ea6a6c7
    feat(server): add logits watermark (#90) OlivierDehaene 2023-03-02 12:30:41 +0100
  • f874c47831
    feat(router): add api-inference headers (#91) OlivierDehaene 2023-03-02 11:41:51 +0100
  • 66da36553a feat(router): add api-inference headers OlivierDehaene 2023-03-02 11:40:45 +0100
  • 0be0506a7e add option to set the watermark gamma & delta from the launcher OlivierDehaene 2023-03-01 17:14:44 +0100
  • 299f7367a5 fix test OlivierDehaene 2023-02-28 18:21:51 +0100
  • 1064950462 feat(server): add logits watermark OlivierDehaene 2023-02-28 17:40:35 +0100
  • a874a52afb sha-4e685d9 OlivierDehaene 2023-02-28 11:24:55 +0100
  • 4e685d907e
    feat(router): ask hf.co for pipelinetag to decide on compat_return_full_text (#89) OlivierDehaene 2023-02-28 10:19:32 +0100
  • d845e377fd fmt OlivierDehaene 2023-02-27 19:34:33 +0100
  • ab32074fe2 better docs OlivierDehaene 2023-02-27 19:30:51 +0100
  • 51b029b089 fmt OlivierDehaene 2023-02-27 19:28:34 +0100
  • f3f9faca2f add return_full_text support OlivierDehaene 2023-02-27 19:22:09 +0100
  • ed22912676 feat(router): add support for return_full_text OlivierDehaene 2023-02-27 17:08:16 +0100
  • 1cddcbbc26 ask the hub OlivierDehaene 2023-02-27 18:19:29 +0100
  • 9579bf165a feat(router): ask hf.co for pipeline to make informed desicion on compat_return_full_text OlivierDehaene 2023-02-27 18:11:49 +0100
  • 21340f24ba
    feat(router): add legacy route for api-inference support (#88) OlivierDehaene 2023-02-27 14:56:58 +0100
  • 91b3a2e2b5 use compat instead of legacy OlivierDehaene 2023-02-27 14:13:39 +0100
  • ca955b810e feat(router): add legacy route for api-inference support OlivierDehaene 2023-02-27 10:31:56 +0100
  • 65e2f1624e
    fix(server): fix token_is_special (#87) OlivierDehaene 2023-02-24 17:20:00 +0100
  • ff7774f2f1 fix(server): fix token_is_special OlivierDehaene 2023-02-24 16:56:34 +0100
  • 3b03c4ea18
    fix(docs): fix openapi schema (#86) OlivierDehaene 2023-02-24 15:59:49 +0100
  • bac7d1c4a3 fix(docs): fix openapi schema OlivierDehaene 2023-02-24 15:59:29 +0100
  • 0ac184ce77
    feat(server): add special token bool (#85) OlivierDehaene 2023-02-24 15:55:57 +0100
  • 4698368a1a revert to batch_decode OlivierDehaene 2023-02-24 15:34:20 +0100
  • ed59f16b96 feat(server): add special token bool OlivierDehaene 2023-02-24 15:22:59 +0100
  • 4b1c9720c0
    v0.3.1 (#84) v0.3.1 OlivierDehaene 2023-02-24 13:27:41 +0100
  • d392ae8333 v0.3.1 OlivierDehaene 2023-02-24 12:52:35 +0100
  • 44ce098c10
    feat(server): pre-allocate max attention mask (#75) OlivierDehaene 2023-02-24 12:49:21 +0100
  • 60ed7b535c first tests megatron Thomas Wolf 2023-02-23 09:52:17 +0100
  • 78063c0569
    fix(server): remove position_ids from galactica forward (#82) OlivierDehaene 2023-02-20 19:28:57 +0100
  • 0363b67d75 fix(server): remove position_ids from galactica forward OlivierDehaene 2023-02-20 19:04:49 +0100
  • f2f78e17d1 better implem OlivierDehaene 2023-02-18 17:30:45 +0100
  • a8446a5a31 feat(server): pre-allocate max attention mask OlivierDehaene 2023-02-17 22:34:36 +0100
  • 17bc841b1b
    feat(server): enable hf-transfer (#76) OlivierDehaene 2023-02-18 14:04:11 +0100
  • c510c30a17 enable hf-transfer even with num-shard==1 OlivierDehaene 2023-02-18 13:29:33 +0100
  • 3bf5c2dd65 feat(server): enable hf-transfer OlivierDehaene 2023-02-18 13:16:18 +0100
  • 6796d38c6d
    feat(router): add cors allow origin options (#73) OlivierDehaene 2023-02-17 18:22:00 +0100
  • 19758f902c docs OlivierDehaene 2023-02-17 18:05:34 +0100
  • 4966a8c8cf feat(router): add cors allow origin options OlivierDehaene 2023-02-17 17:50:43 +0100
  • c720555adc
    v0.3.0 (#72) v0.3.0 OlivierDehaene 2023-02-16 17:28:29 +0100
  • 39f7cd67d3 v0.3.0 OlivierDehaene 2023-02-16 17:28:04 +0100
  • 439fcaf810
    feat(router): add prometheus metrics scrape endpoint (#71) OlivierDehaene 2023-02-16 17:18:53 +0100
  • ef4e0c8247 feat(router): add prometheus metrics scrape endpoint OlivierDehaene 2023-02-16 16:54:07 +0100
  • 7b3d460d21
    fix(launcher): copy current env vars to subprocesses (#70) OlivierDehaene 2023-02-16 11:20:23 +0100
  • 14f1818ba1 fix(launcher): copy current env vars to subprocesses OlivierDehaene 2023-02-16 10:53:05 +0100
  • 5437d49beb
    feat(router): add max_total_tokens and empty_input validation (#68) OlivierDehaene 2023-02-15 21:56:59 +0100
  • 32d7e5e20f remove useless max_max_new_tokens OlivierDehaene 2023-02-15 21:34:40 +0100
  • d6fc264a99 typo OlivierDehaene 2023-02-15 21:19:41 +0100
  • bfdd8de903 feat(router): add max_total_tokens and empty_input validation OlivierDehaene 2023-02-15 21:18:38 +0100
  • 68455353f5
    feat(launcher): add disable_custom_kernels arg (#67) OlivierDehaene 2023-02-15 16:23:45 +0100
  • c5a4a1faf3
    feat(server): improve download logging (#66) OlivierDehaene 2023-02-15 16:11:32 +0100
  • 72a776ca56 feat(launcher): add disable_custom_kernels arg OlivierDehaene 2023-02-15 16:04:24 +0100
  • e77d9b86ea small changes OlivierDehaene 2023-02-15 15:52:11 +0100
  • 603a1b3f48 feat(server): improve download logging OlivierDehaene 2023-02-15 15:47:12 +0100
  • 0fbc691946
    feat: add safetensors conversion (#63) OlivierDehaene 2023-02-14 13:02:16 +0100
  • 9034105553 update launcher OlivierDehaene 2023-02-14 12:35:36 +0100
  • 7a0bbf0994 improve readme OlivierDehaene 2023-02-14 12:13:38 +0100
  • 97f9ae6a6d let launcher download weights OlivierDehaene 2023-02-14 12:09:58 +0100
  • 975bbda03b
    Add note about NVIDIA drivers lewtun 2023-02-14 12:05:04 +0100
  • 397a28080c feat: add safetensors conversion OlivierDehaene 2023-02-13 16:13:04 +0100
  • 9af454142a
    feat: add distributed tracing (#62) OlivierDehaene 2023-02-13 13:02:45 +0100
  • ca73b60da7 remove max sequence length OlivierDehaene 2023-02-13 12:46:39 +0100
  • 3fa6a4e674 fix seq2seq OlivierDehaene 2023-02-10 19:25:07 +0100
  • e9441a1ea2 use only last logits OlivierDehaene 2023-02-10 18:36:34 +0100
  • 189fba28d1 update readme and dockerfile to use latest version of protoc OlivierDehaene 2023-02-10 15:49:09 +0100
  • e39302c3e5 add sudo OlivierDehaene 2023-02-10 15:36:52 +0100
  • 26c7bdeab2 formatting OlivierDehaene 2023-02-10 15:35:01 +0100
  • 67cd625c82 improved instrumentation OlivierDehaene 2023-02-10 15:30:53 +0100
  • f81f0828d7 cleanup OlivierDehaene 2023-02-10 14:04:58 +0100
  • 1e5a30990b Update dependencies OlivierDehaene 2023-02-10 12:14:29 +0100
  • 7fa81a05b0 add shutdown procedure OlivierDehaene 2023-02-09 19:57:53 +0100
  • b3cc379550 add tracing to rust router OlivierDehaene 2023-02-09 19:24:09 +0100
  • 04015dfa90 feat: add distributed tracing OlivierDehaene 2023-02-09 15:07:08 +0100
  • 6aa17d7923 sha-e520d5b OlivierDehaene 2023-02-09 11:14:36 +0100
  • e520d5b349
    fixed SSE naming (#61) Yannic Kilcher 2023-02-08 22:30:11 +0100
  • 6d22c7c92f fixed SSE naming Yannic Kilcher 2023-02-08 21:43:14 +0100
  • 1ad3250b89
    fix(docker): increase shm size (#60) OlivierDehaene 2023-02-08 17:53:33 +0100
  • 8acf05dc89 fix(docker): increase shm size OlivierDehaene 2023-02-08 17:28:54 +0100
  • c503a639b1
    feat(server): support t5 (#59) OlivierDehaene 2023-02-07 18:25:17 +0100
  • a96b4cdbe2 feat(server): support t5 OlivierDehaene 2023-02-07 18:21:26 +0100
  • 645e0a3878 v0.2.1 OlivierDehaene 2023-02-07 16:30:06 +0100
  • 2fe5e1b30e
    V0.2.1 (#58) v0.2.1 OlivierDehaene 2023-02-07 15:40:25 +0100
  • d48af7feb8 update cargo.lock OlivierDehaene 2023-02-07 15:39:44 +0100
  • 04558837dd v0.2.1 OlivierDehaene 2023-02-07 15:38:47 +0100
  • 4acc42a605
    fix(server): better handling of inference mode (#57) OlivierDehaene 2023-02-07 15:38:22 +0100
  • 5fb826ca14 better handling of inference mode OlivierDehaene 2023-02-07 15:23:20 +0100
  • d8b84cc025 feat(server): modify nccl init_method OlivierDehaene 2023-02-07 14:38:43 +0100
  • 80d03723a7 v0.2.0 OlivierDehaene 2023-02-06 18:11:23 +0100
  • e114d87486
    feat(ci): push to AML registry (#56) OlivierDehaene 2023-02-06 14:33:56 +0100
  • 3b06d36c1c feat(ci): push to AML registry OlivierDehaene 2023-02-06 14:17:26 +0100
  • a0dca443dd
    feat(docs): Clarify installation steps (#54) lewtun 2023-02-03 13:07:55 +0100
  • bc8596390c Clarify installation steps lewtun 2023-02-03 12:24:20 +0100
  • 20c3c5940c
    feat(router): refactor API and add openAPI schemas (#53) v0.2.0 OlivierDehaene 2023-02-03 12:43:37 +0100