huggingface/text-generation-inference

Pull Requests Commits
fix: lint backend and doc files (#2850)
drbh committed 1 year ago
Verified a72f339c
Fixing CI. (#2846)
Narsil committed 1 year ago
Verified 11ab3298
New arg. (#2845)
Narsil committed 1 year ago
Verified 6f0b8c94
Feat/trtllm cancellation dev container (#2795)
Hugoch committed 1 year ago
Verified 1708865f
TensorRT-LLM backend bump to latest version + misc fixes (#2791)
mfuntowicz committed 1 year ago
Verified ea7f4082
Fixup opt to reduce the amount of odd if statements. (#2833)
Narsil committed 1 year ago
Verified 3bb3fd19
fix facebook/opt-125m not working issue (#2824)
sywangyi committed 1 year ago
Verified bf59118a
Fixing latest flavor by disabling it. (#2831)
Narsil committed 1 year ago
Verified c3bd7212
docs(README): supported hardware links TGI AMD GPUs (#2814)
guspan-tanadi committed 1 year ago
Verified f01f2fb6
Prepare patch release. (#2829)
Narsil committed 1 year ago
Verified 07b01293
Update README.md (#2827)
RodriMora committed 1 year ago
Verified cc66dccb
Using both value from config as they might not be correct. (#2817)
Narsil committed 1 year ago
Verified 82c24f74
Small update to docs (#2816)
Narsil committed 1 year ago
Verified a2d878fa
Hotfix link2 (#2812)
Narsil committed 1 year ago
Verified b2fac5d9
Hotfixing the link. (#2811)
Narsil committed 1 year ago
Verified a70dd299
Prep new version (#2810)
Narsil committed 1 year ago
Verified 042791fb
V3 doc (#2809)
Narsil committed 1 year ago
Verified 27fa83ca
Attempt for cleverer auto batch_prefill values (some simplifications). (#2808)
Narsil committed 1 year ago
Verified a04356fb
Enable paligemma2 (#2807)
drbh committed 1 year ago
Verified 9f5c9a5e
Removing experimental to prefill chunking.
Narsil committed 1 year ago
Verified 08f6fa0b
Adding A100 compute. (#2806)
Narsil committed 1 year ago
Verified d96dcb17
Auto max prefill (#2797)
Narsil committed 1 year ago
Verified 5df80590
feat: auto max_new_tokens (#2803)
OlivierDehaene committed 1 year ago
Verified 8c3669b2
use oneapi 2024 docker image directly for xpu (#2793)
sywangyi committed 1 year ago
Verified 6685e8fc
fix: avoid setting use_sgmv if no kernels present (#2796)
drbh committed 1 year ago
Verified e0db6333
Saving some VRAM. (#2790)
Narsil committed 1 year ago
Verified b57f3703
Sync (most) server dependencies with Nix (#2782)
danieldk committed 1 year ago
Verified 2003d8be
fix: only use eos_token_id as pad_token_id if int (#2774)
dvrogozh committed 1 year ago
Verified 535149d8
fix: add merge-lora arg for model id (#2788)
drbh committed 1 year ago
Verified 2c74c556
Removing ../ that broke the link (#2789)
Getty committed 1 year ago
Verified a35d1e6f
Terms Of Use Privacy Policy Refund Policy Impressum