Default Branch

09e9068007 · whisper.android : support benchmark for Android example. (#542) · Updated 2023-03-07 20:36:30 +01:00

Branches

4f074fb7a8 · tmp : demonstrate how to measure time of ggml ops · Updated 2023-03-09 08:28:06 +01:00

0
1

bf5d4c81b9 · make : fix MUSL Linux build · Updated 2023-03-06 21:24:08 +01:00

2
1

17a14593de · coreml : simlpify whisper_encode + log messages · Updated 2023-03-05 17:31:09 +01:00

8
2

b4ebdb6b57 · bench : add Q4_0 and Q4_1 mul_mat benchmarks · Updated 2023-02-27 19:14:32 +01:00

13
5

a0da7f71a2 · command : wip in progress, improve guided decoding · Updated 2023-02-19 18:39:05 +01:00

16
1

ec44ad0a75 · diarization : try conv and self-attention embeddings · Updated 2023-02-19 12:00:12 +01:00

17
4

59c997ca2d · wip ignore · Updated 2023-02-15 18:11:12 +01:00

24
1

7aa1174315 · bench : fix Windows linkage by moving ggml benches in whisper lib .. · Updated 2023-01-18 20:16:25 +01:00

62
1

e2aa556a99 · whisper : experiments with Flash Attention in the decoder · Updated 2023-01-07 20:00:51 +01:00

84
1

4e6d2e98ab · ggml : try to improve threading · Updated 2022-12-29 12:05:20 +01:00

124
1

683f111088 · ggml : initial tests with libnvblas · Updated 2022-12-08 21:01:52 +01:00

203
1

e0bd97f41f · ggml : use macros to inline FP16 <-> FP32 conversions · Updated 2022-12-06 21:05:33 +01:00

207
1

0a2621b637 · stream : add "max_tokens" cli arg · Updated 2022-11-20 20:22:02 +01:00

285
5

fa9621e5e9 · mtl : update Makefile to support Metal · Updated 2022-11-12 07:32:03 +01:00

295
2

5d895d60b6 · Merge branch 'master' into avx512 · Updated 2022-11-06 08:09:50 +01:00

300
7

210a6fb83c · wip : some unsuccessful experiments using DP · Updated 2022-11-01 20:28:30 +01:00

323
1

4597c9c19b · wip : try to compress just mlp · Updated 2022-10-08 14:12:15 +02:00

437
2