[Continuous batching] Add finish reason to generation output #1623
causal_lm_cpp.yml
on: pull_request
cpp-multinomial-greedy_causal_lm-ubuntu
7m 43s
cpp-greedy_causal_lm-windows
18m 55s
cpp-beam_search_causal_lm-Qwen-7B-Chat
11m 46s
cpp-beam_search_causal_lm-Qwen1_5-7B-Chat
14m 58s
cpp-beam_search_causal_lm-Phi-2
7m 21s
cpp-beam_search_causal_lm-notus-7b-v1
9m 2s
cpp-speculative_decoding_lm-ubuntu
9m 48s
cpp-prompt_lookup_decoding_lm-ubuntu
7m 2s
cpp-Phi-1_5
6m 25s
cpp-greedy_causal_lm-redpajama-3b-chat
13m 47s
cpp-chat_sample-ubuntu
9m 57s
cpp-continuous-batching-ubuntu
7m 4s
cpp-continuous-batching-windows
18m 9s
cpp-continuous-batching-macos
12m 28s
Matrix: cpp-beam_search_causal_lm-ubuntu
Annotations
16 warnings