Commit Graph

  • 0fec9fbfd1 feat(python-client): release v0.4.0 OlivierDehaene 2023-03-23 18:07:02 +0100
  • 5e5e9d4bbd
    feat: Add note about NVIDIA drivers (#64) lewtun 2023-03-23 18:03:45 +0100
  • c07acd4fea
    Merge branch 'main' into lewtun-patch-1 OlivierDehaene 2023-03-23 18:03:33 +0100
  • 603e20b5f7
    feat(ci): add ci paths (#134) OlivierDehaene 2023-03-23 18:01:30 +0100
  • 7850119055
    feat(python-client): add cookies to Client constructors and requests (#132) dconathan 2023-03-23 13:01:01 -0400
  • 7fcbab25a0 feat(ci): add ci paths OlivierDehaene 2023-03-23 18:00:28 +0100
  • a87468ad86 black OlivierDehaene 2023-03-23 17:51:30 +0100
  • d199c71a32 make neox go brrr OlivierDehaene 2023-03-23 17:47:15 +0100
  • 81b5fceb52 fix inference_api headers for new signature Devin Conathan 2023-03-23 09:04:31 -0400
  • a4df5bc64a faster OlivierDehaene 2023-03-23 14:01:35 +0100
  • e5e22993e7 faster OlivierDehaene 2023-03-23 13:33:32 +0100
  • 19a04f22dd pre-compute OlivierDehaene 2023-03-23 13:10:31 +0100
  • cdc70f4c23 faster OlivierDehaene 2023-03-23 12:35:38 +0100
  • ead19abb0e faster OlivierDehaene 2023-03-23 12:00:52 +0100
  • 232bfbf27f add cookies to Client constructors and requests Devin Conathan 2023-03-22 22:31:44 -0400
  • 24579c45de wip OlivierDehaene 2023-03-22 11:46:09 +0100
  • 2ae7c337a1 added llama Yannic Kilcher 2023-03-19 09:31:47 +0100
  • a3b7db932f
    fix(python-client): relax dependencies (#129) OlivierDehaene 2023-03-16 12:57:07 +0100
  • 9c5f56986a fix(python-client): relax dependencies OlivierDehaene 2023-03-16 12:56:36 +0100
  • b49dbf2d88
    fix(server): use server tokenizer as gt (#128) OlivierDehaene 2023-03-16 12:12:26 +0100
  • 9ef62c0209 fix(server): use server tokenizer as gt OlivierDehaene 2023-03-16 11:50:53 +0100
  • 8ad60b752f
    fix(server): add position ids to neox (#126) OlivierDehaene 2023-03-15 13:12:49 +0100
  • 9befecdd7d fix(server): add position ids to neox OlivierDehaene 2023-03-15 10:42:13 +0100
  • cbd36aa4d1
    fix(server): revert gpt-neox optims (#123) OlivierDehaene 2023-03-13 22:57:08 +0100
  • 13b6f2cacf fix(server): revert gpt-neox optims OlivierDehaene 2023-03-13 22:56:44 +0100
  • 6860ce9c67
    feat: add OpenAssistant/oasst-sft-1-pythia-12b to the list of supported models (#122) OlivierDehaene 2023-03-13 20:42:10 +0100
  • f4421bab79 feat: add OpenAssistant/oasst-sft-1-pythia-12b to the list of supported models OlivierDehaene 2023-03-13 20:41:46 +0100
  • 47ac334a21 0.4.0 deploy/aml OlivierDehaene 2023-03-12 10:06:15 +0100
  • c01d9b9d99 revert to old version OlivierDehaene 2023-03-10 14:39:35 +0100
  • 411d6247f4
    v0.4.0 (#119) v0.4.0 OlivierDehaene 2023-03-09 16:07:01 +0100
  • 98ea40519c v0.4.0 OlivierDehaene 2023-03-09 16:05:13 +0100
  • d8dc8f1b0c
    feat(python-client): add new parameters (#118) OlivierDehaene 2023-03-09 16:05:33 +0100
  • 0d1a8fc250 test best_of OlivierDehaene 2023-03-09 16:02:37 +0100
  • 55bd4fed7d
    feat(router): add best_of parameter (#117) OlivierDehaene 2023-03-09 15:30:54 +0100
  • becec0d501 update schema OlivierDehaene 2023-03-09 15:15:51 +0100
  • a448acbfbe fmt OlivierDehaene 2023-03-09 15:11:16 +0100
  • 8d7a0c1992 force sampling when using best_of OlivierDehaene 2023-03-09 14:50:42 +0100
  • 9f4f2fc8e3 add best of sequences to details OlivierDehaene 2023-03-09 14:27:39 +0100
  • 9624d4060f docstring OlivierDehaene 2023-03-09 13:09:11 +0100
  • 5932ff4aa2 feat(router): add best_of parameter OlivierDehaene 2023-03-09 13:06:12 +0100
  • 1990d8633c add validation OlivierDehaene 2023-03-09 15:04:59 +0100
  • c5a0b65c47 use flan-t5 for tests OlivierDehaene 2023-03-09 14:39:00 +0100
  • 5e1473f0f8 feat(python-client): add new parameters OlivierDehaene 2023-03-09 13:48:58 +0100
  • e8bfe199ba
    feat(router): support left truncation (#115) OlivierDehaene 2023-03-09 13:10:30 +0100
  • c0795de2f2
    fix(server): do not warp prefill logits (#116) OlivierDehaene 2023-03-09 13:00:10 +0100
  • d405880504 support left truncate OlivierDehaene 2023-03-09 11:09:36 +0100
  • a376d8bc59 wip OlivierDehaene 2023-03-09 10:38:11 +0100
  • 1a2d68250a
    feat: support typical sampling (#114) OlivierDehaene 2023-03-09 11:33:57 +0100
  • f49786ccba fix(server): do not warp prefill logits OlivierDehaene 2023-03-09 11:33:28 +0100
  • 6def47158f fix test OlivierDehaene 2023-03-09 11:13:36 +0100
  • 864766c89b add default value OlivierDehaene 2023-03-09 10:30:26 +0100
  • 05ad316448 fmt OlivierDehaene 2023-03-09 10:17:50 +0100
  • 140285c1f7 feat: support typical sampling OlivierDehaene 2023-03-09 10:17:18 +0100
  • c3e2b79a9e update probes OlivierDehaene 2023-03-09 09:52:11 +0100
  • 941cd42e0c
    fix(server): fix index out of range for watermarking (#110) OlivierDehaene 2023-03-08 18:29:08 +0100
  • 4abf27ce81 remove vocab size OlivierDehaene 2023-03-08 18:17:34 +0100
  • 7a6a7ed27b black OlivierDehaene 2023-03-08 17:57:54 +0100
  • 5dbfc07c6e fix(server): fix index out of range for watermarking OlivierDehaene 2023-03-08 17:55:21 +0100
  • 2c5df5d2af
    fix(python-client): stream not set on the sync client (#109) OlivierDehaene 2023-03-08 16:48:16 +0100
  • 6a6444e7a0 fix readme OlivierDehaene 2023-03-08 16:47:39 +0100
  • 1bc6742949 fix(python-client): stream not set OlivierDehaene 2023-03-08 16:46:50 +0100
  • 5fd2dcb513
    feat(launcher): default num_shard to CUDA_VISIBLE_DEVICES if possible (#108) OlivierDehaene 2023-03-08 13:53:41 +0100
  • 33e8565738 add validation OlivierDehaene 2023-03-08 13:11:25 +0100
  • 2896a7c410 log OlivierDehaene 2023-03-08 13:04:02 +0100
  • 466963238c fmt OlivierDehaene 2023-03-08 13:03:19 +0100
  • b761d02713 feat(launcher): default num_shard to CUDA_VISIBLE_DEVICES if possible OlivierDehaene 2023-03-08 13:02:17 +0100
  • 0ac38d336a
    feat(launcher): allow parsing num_shard from CUDA_VISIBLE_DEVICES (#107) OlivierDehaene 2023-03-08 11:06:59 +0100
  • ea634ad693 update package version OlivierDehaene 2023-03-08 10:54:18 +0100
  • 0d8b30d19a fmt OlivierDehaene 2023-03-08 10:51:17 +0100
  • e4142e4fd5 feat(launcher): allow parsing num_shard from CUDA_VISIBLE_DEVICES OlivierDehaene 2023-03-08 10:49:18 +0100
  • b1485e18c5
    fix(server): fix galactica batch (#106) OlivierDehaene 2023-03-07 20:05:21 +0100
  • 97d1c68772 fix(server): fix galactica batch OlivierDehaene 2023-03-07 20:04:49 +0100
  • 3fef90d50f
    feat(clients): Python client (#103) OlivierDehaene 2023-03-07 18:52:22 +0100
  • a3e57d3c5d update readme OlivierDehaene 2023-03-07 18:43:00 +0100
  • 478d5c1403 publish OlivierDehaene 2023-03-07 18:13:34 +0100
  • 6e9e194f33 final OlivierDehaene 2023-03-07 17:55:23 +0100
  • b7c3c7dabb wrong import OlivierDehaene 2023-03-07 13:57:32 +0100
  • f3586b2308 fmt OlivierDehaene 2023-03-07 13:35:37 +0100
  • feddbbc998 poc OlivierDehaene 2023-03-07 13:33:38 +0100
  • 0e9ed1a8c2
    feat: add supported models (#102) OlivierDehaene 2023-03-07 12:55:05 +0100
  • ea0a8078fe add newline OlivierDehaene 2023-03-07 12:54:43 +0100
  • 8390d6bdb3 feat: add supported models OlivierDehaene 2023-03-07 12:53:59 +0100
  • 6d1ad06471 wip OlivierDehaene 2023-03-07 12:53:03 +0100
  • f067673a1d increased initial delay OlivierDehaene 2023-03-07 11:14:31 +0100
  • 0a27d56634 wip OlivierDehaene 2023-03-07 10:14:49 +0100
  • 31159266eb sha-cd5961b OlivierDehaene 2023-03-06 18:05:19 +0100
  • c543b9c585 wip OlivierDehaene 2023-03-06 16:31:15 +0100
  • cd5961b5da
    feat: allow local models (#101) OlivierDehaene 2023-03-06 14:39:36 +0100
  • 590fc3794c remove unused name checks OlivierDehaene 2023-03-06 13:48:07 +0100
  • 02df3dea9d feat: allow local models OlivierDehaene 2023-03-06 13:40:47 +0100
  • 9b205d33cc
    fix(server): fix generate_stream by forcing tokens to be decoded correctly (#100) OlivierDehaene 2023-03-06 13:22:58 +0100
  • 56d23753bb add clean_up_tokenization_spaces OlivierDehaene 2023-03-06 13:10:12 +0100
  • 6a56f945c0 fix(server): fix generate_stream by forcing tokens to be decoded correctly OlivierDehaene 2023-03-06 12:59:54 +0100
  • 1c19b0934e
    v0.3.2 (#97) v0.3.2 OlivierDehaene 2023-03-03 18:42:20 +0100
  • f50fdf25c3 v0.3.2 OlivierDehaene 2023-03-03 18:24:05 +0100
  • 0b6807caa4
    feat(server): fix transformers commit (#96) OlivierDehaene 2023-03-03 17:56:27 +0100
  • c02a5475ea enable hf_transfer on tests OlivierDehaene 2023-03-03 17:44:50 +0100
  • 5862add4d0 feat(server): fix transformers commit OlivierDehaene 2023-03-03 17:33:35 +0100
  • 240c4187fd
    fix(launcher): add router parameters to launcher (#95) OlivierDehaene 2023-03-03 16:01:25 +0100
  • 11e9a99046 fix(launcher): add router parameters to launcher OlivierDehaene 2023-03-03 15:44:56 +0100