huggingface/text-generation-inference

Pull Requests Commits
Update __init__.py (#794)
Narsil committed 2 years ago
Verified c4dac9f3
Fixing non 4bits quantization. (#785)
Narsil committed 2 years ago
Verified 1fdc88ee
Fix dynamic rope. (#783)
Narsil committed 2 years ago
Verified 891e19cc
Merge BNB 4bit. (#770)
Narsil committed 2 years ago
Verified 16fadcec
fix build tokenizer in quantize and remove duplicate import (#768)
zspo committed 2 years ago
Verified f91e9d28
This should prevent the PyTorch overriding. (#767)
Narsil committed 2 years ago
Verified 6ec5288a
feat(server): Add native support for PEFT Lora models (#762)
Narsil committed 2 years ago
Verified ac736fd8
Automatically map deduplicated safetensors weights to their original values (#501) (#761)
Narsil committed 2 years ago
Verified 8b0d608f
add FastLinear import (#750)
zspo committed 2 years ago
Verified bd308874
fix typo for dynamic rotary (#745)
flozi00 committed 2 years ago
Verified 7766fee9
Typo fix. (#746)
Narsil committed 2 years ago
Verified d3d8f1bd
fix(server): Failing quantize config after local read. (#743)
Narsil committed 2 years ago
Verified 15fc6466
chore: fix typo in mpt_modeling.py (#737)
eltociear committed 2 years ago
Verified 2a13f1a0
Adding Rope scaling. (#741)
Narsil committed 2 years ago
Verified 932bdd93
Fix typing in `Model.generate_token` (#733)
jaywonchung committed 2 years ago
Verified b9633c46
Local gptq support. (#738)
Narsil committed 2 years ago
Verified 92bb56b0
v1.0.0 (#727)
OlivierDehaene committed 2 years ago
Verified 3ef5ffbc
chore: update license to HFOIL (#725)
OlivierDehaene committed 2 years ago
Verified bde25e62
feat(server): update vllm version (#723)
OlivierDehaene committed 2 years ago
Verified afd04dc7
docs: Add hardware section to TOC in README (#721)
regisss committed 2 years ago
Verified f848dece
Add section about TGI on other AI hardware accelerators in README (#715)
regisss committed 2 years ago
Verified 5a1cccbb
v0.9.4 (#713)
OlivierDehaene committed 2 years ago
Verified 9f18f4c0
feat(server): support new falcon config (#712)
OlivierDehaene committed 2 years ago
Verified ab96b9ae
fix(server): fix missing datasets in quantize
OlivierDehaene committed 2 years ago
2efd46ef
fix(server): fix quantization python requirements (#708)
OlivierDehaene committed 2 years ago
Verified 8bd0adb1
docs(README): update readme
OlivierDehaene committed 2 years ago
e64a6589
feat(server): Using `quantize_config.json` instead of GPTQ_BITS env variables. (#671)
Narsil committed 2 years ago
Verified a0d55358
fix(server): fix exllama buffers (#689)
OlivierDehaene committed 2 years ago
Verified 37df6df3
feat: add cuda memory fraction (#659)
OlivierDehaene committed 2 years ago
Verified 73a4d65d
feat(server): add local prom and health routes if running w/ ngrok
OlivierDehaene committed 2 years ago
1da642bd
Terms Of Use Privacy Policy Refund Policy Impressum