huggingface/text-generation-inference

Pull Requests Commits

Pre-emptive on sealion.

Narsil committed 1 year ago

45978034

Update the docs

Narsil committed 1 year ago

17b7b75e

fix: read stderr in download (#1486)

OlivierDehaene committed 1 year ago

Verified 9c320e26

feat: adds phi model (#1442)

drbh committed 1 year ago

Verified 7e2a7433

Add a new `/tokenize` route to get the tokenized input (#1471)

Narsil committed 1 year ago

Verified 86c8335f

Add messages api compatibility docs (#1478)

drbh committed 1 year ago

Verified 7872b8c5

Fixing non divisible embeddings. (#1476)

Narsil committed 1 year ago

Verified 7e542d4d

Disable `decoder_input_details` on OpenAI-compatible chat streaming, pass temp and top-k from API (#1470)

EndlessReform committed 1 year ago

Verified 82f87ada

feat: conditionally toggle chat on invocations route (#1454)

drbh committed 1 year ago

Verified 98e5faff

chore: bump rust version and annotate/fix all clippy warnings (#1455)

drbh committed 1 year ago

Verified becd0997

feat: support raise_exception, bos and eos tokens (#1450)

drbh committed 1 year ago

Verified 3ccb3bb0

feat: supports openai chat completions API (#1427)

drbh committed 1 year ago

Verified 0eabc835

Return prompt vs generated tokens. (#1436)

Narsil committed 1 year ago

Verified ac08b4ef

Fix local load for Medusa (#1420)

PYNing committed 1 year ago

Verified da27fbdf

fix: follow base model for tokenizer in router (#1424)

OlivierDehaene committed 1 year ago

Verified fbeb1c44

Fix missing make target platform for local install: 'install-flash-attention-v2' (#1414)

deepily committed 1 year ago

Verified 91d72675

fix: fix local loading for .bin models (#1419)

OlivierDehaene committed 1 year ago

Verified 564f2a3b

docs: update required CUDA version to 12.2

OlivierDehaene committed 1 year ago

3f9b3f45

OlivierDehaene committed 1 year ago

630800ee

Fix local load for peft (#1373)

Narsil committed 1 year ago

Verified 529d7c25

feat: update exllamav2 kernels (#1370)

OlivierDehaene committed 1 year ago

Verified 564199ba

docs: Change URL for Habana Gaudi support in doc (#1343)

regisss committed 1 year ago

Verified 987c959f

Peft safetensors. (#1364)

Narsil committed 1 year ago

Verified eb8923a9

fix: fix gpt-q with groupsize = -1 (#1358)

OlivierDehaene committed 2 years ago

Verified d077150e

fix: fix offline (#1341) (#1347)

OlivierDehaene committed 2 years ago

Verified 8428ed10

fix: fix logic if sliding window key is not present in config (#1352)

OlivierDehaene committed 2 years ago

Verified 1b1bfa49

feat: relax mistral requirements (#1351)

OlivierDehaene committed 2 years ago

Verified 9b56d3fb

OlivierDehaene committed 2 years ago

f3aea78f

fix: max_past default value must be -1, not 0 (#1348)

OlivierDehaene committed 2 years ago

Verified 37555cf4

fix: only keep stop sequence buffer if we have some

OlivierDehaene committed 2 years ago

9b78a6ee

Older