text-generation-inference
Trtllm backend improvements
#3231
Open
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
13
Changes
View On
GitHub
Commits
feat(trtllm): add new finish reasons
leejuyuu
committed
71 days ago
fix: fix prometheus_port CLI short arg conflict
leejuyuu
committed
71 days ago
fix(trtllm): fix segfault when canceling request
leejuyuu
committed
71 days ago
feat(trtllm): add stop sequence support
leejuyuu
committed
71 days ago
feat(trtllm): catch broader exception
leejuyuu
committed
71 days ago
feat(trtllm): check existence of config files
leejuyuu
committed
71 days ago
fix(trtllm): fix do_sample being ignored
leejuyuu
committed
71 days ago
feat(trtllm): get more accurate start time
leejuyuu
committed
71 days ago
perf(trtllm): reduce futile loop iterations
leejuyuu
committed
71 days ago
refactor: add interior mutability to tensorrt_llm_backend_t
leejuyuu
committed
71 days ago
feat(trtllm): separate request and response loop
leejuyuu
committed
71 days ago
fix(trtllm): handle single eos_token_id in generation_config
leejuyuu
committed
71 days ago
feat(trtllm): support guided decoding
leejuyuu
committed
71 days ago
Loading