为了安全，强烈建议开启2FA双因子认证：User Settings -> Account -> Enable two-factor authentication！！！

Tags

Tags give the ability to mark specific points in history as being important

b2817

83330d8c · main : add --conversation / -cnv flag (#7108) · May 08, 2024
b2816

465263d0 · sgemm : AVX Q4_0 and Q8_0 (#6891) · May 08, 2024
b2815

911b3900 · server : add_special option for tokenize endpoint (#7059) · May 08, 2024
b2813

229ffff8 · llama : add BPE pre-tokenization for Qwen2 (#7114) · May 08, 2024
b2812

1fd9c174 · clean up json_value & server_log (#7142) · May 08, 2024
b2811

4cd621c2 · convert : add BPE pre-tokenization for DBRX (#7132) · May 08, 2024
b2809

acdce3cd · compare-llama-bench.py: add missing basicConfig (#7138) · May 08, 2024
b2808

38554160 · ggml : introduce bfloat16 support (#6412) · May 08, 2024
b2806

c780e753 · Further tidy on Android instructions README.md (#7077) · May 08, 2024
b2805

48b2f9c1 · Fixed save_imatrix to match old behaviour for MoE (#7099) · May 08, 2024
b2804

af0a5b61 · server: fix incorrectly reported token probabilities (#7125) · May 07, 2024
b2803

b6aa6702 · Fix OLMo HF to GGUF conversion (#6910) · May 07, 2024
b2800

3af34c1d · main : update log text (EOS to EOG) (#7104) · May 07, 2024
b2797

858f6b73 · Add an option to build without CUDA VMM (#7067) · May 06, 2024
b2794

628b2991 · Adding support for the --numa argument for llama-bench. (#7080) · May 05, 2024
b2793

8f8acc86 · Disable benchmark on forked repo (#7034) · May 05, 2024
b2791

889bdd76 · command-r : add BPE pre-tokenization (#7063) · May 05, 2024
b2789

84250014 · gguf-split: add --no-tensor-first-split (#7072) · May 04, 2024
b2787

fcd84a0f · Fix Linux /sys cpu path to guess number of cores (#7064) · May 04, 2024
b2786

03fb8a00 · If first token generated from the server is the stop word the server will crash (#7038) · May 04, 2024

1
…
83
84
85
86
87
88
89
90
91
…
178