Skip to content
GitLab
Explore
Sign in
Overview
Active
Stale
All
This project is mirrored from
https://github.com/ggerganov/whisper.cpp
. Pull mirroring updated
Sep 19, 2024
.
experiment/model-compression
4597c9c1
·
wip : try to compress just mlp
·
Oct 08, 2022
word-ts-2
210a6fb8
·
wip : some unsuccessful experiments using DP
·
Nov 01, 2022
avx512
5d895d60
·
Merge branch 'master' into avx512
·
Nov 06, 2022
metal
fa9621e5
·
mtl : update Makefile to support Metal
·
Nov 12, 2022
stream
0a2621b6
·
stream : add "max_tokens" cli arg
·
Nov 20, 2022
macros-cvt-fp16
e0bd97f4
·
ggml : use macros to inline FP16 <-> FP32 conversions
·
Dec 06, 2022
nvblas
683f1110
·
ggml : initial tests with libnvblas
·
Dec 08, 2022
threads
4e6d2e98
·
ggml : try to improve threading
·
Dec 29, 2022
fa-decoder
e2aa556a
·
whisper : experiments with Flash Attention in the decoder
·
Jan 07, 2023
arghh
7aa11743
·
bench : fix Windows linkage by moving ggml benches in whisper lib ..
·
Jan 18, 2023
chess
59c997ca
·
wip ignore
·
Feb 15, 2023
diarization
ec44ad0a
·
diarization : try conv and self-attention embeddings
·
Feb 19, 2023
guided
a0da7f71
·
command : wip in progress, improve guided decoding
·
Feb 19, 2023
timing
4f074fb7
·
tmp : demonstrate how to measure time of ggml ops
·
Mar 09, 2023
coreml-with-state
02448106
·
rebase on master after whisper_state changes
·
Mar 26, 2023
talk.llama-coreml
3627ef51
·
minor
·
Mar 26, 2023
llama-podcast
c456ca47
·
llama podcast
·
Apr 01, 2023
fix-vzip
c6174cb8
·
wip
·
May 02, 2023
fix-coreml-ane
8cbc3635
·
coreml : attempt to fix ANE-optimized models
·
Jul 11, 2023
java-bindings
3efe146d
·
ci : enable java package publishing
·
Aug 30, 2023
Prev
1
2
3
Next