Spaces:
Running
Running
Commit History
ggml : sync latest ggml
7b8292f
unverified
ggml : fix q4_1 dot product types (#759)
984a856
unverified
ggml : sync latest changes from ggml and llama.cpp
3bd52ce
unverified
ggml : fix WASM build
70332a0
unverified
ggml : backport llama.cpp updates (close #709)
bf6b4f8
unverified
talk-llama : add new example + sync ggml from llama.cpp (#664)
a8c74e6
unverified
whisper : reduce memory usage during inference (#431)
3aa9e6c
unverified
whisper : PPC64 big-endian support (#398)
239569b
unverified
fitzsim
commited on
bench : add memcpy and ggml_mul_mat benchmarks
a660ed9
unverified
ggml : remove obsolete zeroing + comment fixes (#390)
9c35c0d
unverified
ggml : correct behaviour of ggml_vec_sum_f32 (#390)
ffffc6e
unverified
Abitofevrything
commited on
ggml : improve vec_dot_f16 unrolling in flash_attn_f16
6e57274
unverified
ggml : fix bug in new soft max computation
c59ce76
unverified
ggml : when using BLAS start only 1 CPU thread
6c4692f
unverified
ggml : fix running tasks with variable number of threads
2078d85
unverified
ggml : unroll ggml_vec_dot_f16 in ggml_compute_forward_flash_attn_f16
f07fecd
unverified
whisper : revert accidental MB change
db991e1
unverified
ggml : speed-up soft max via Accelerate + unroll
fdaf59a
unverified
ggml : use vDSP_sve and vDSP_maxv from Accelerate
ed14a8b
unverified
ggml : make gcc happy (minor)
496acd2
unverified
ggml : add SSE3 and fp16 conversion lookup table (#368)
2c3f7d4
unverified
whisper : document POWER VSX support
4dbf7ee
Thomas Fitzsimmons
commited on
ggml : reorganize POWER9 ppc64le SIMD code
e0a5614
Thomas Fitzsimmons
commited on
ggml : change f16 load and store macro arguments
4a68b87
Thomas Fitzsimmons
commited on
ggml : add void to argument-less functions
f06f912
unverified
ggml : define MIN / MAX only if not defined (minor)
2117da6
unverified
ggml : improve f16 acceleration for POWER9 ppc64le
f92a260
Thomas Fitzsimmons
commited on
ggml : barrier refactor + static functions
7b501c1
unverified
ggml : simplify the SIMD code (#324)
6fe850c
unverified
ggml : use vaddvq_f32 for slightly more efficient reduce
550fbf8
unverified
ggml : add f16 acceleration for POWER9 ppc64le
0d5a830
Thomas Fitzsimmons
commited on
ggml : make consts static (#317)
786be65
unverified
Andy Maloney
commited on
minor : small code cleanups (#302)
142f526
unverified
Check for both __ARM_NEON and __ARM_FEATURE_FMA so that the project can be compiled for armv7a.
1fff54f
Kevin Brothaler
commited on
Add AVX,AVX2 support for ggml_vec_scale_f32
666b50a
katsu560
commited on
ggml : implement ggml_compute_forward_dup_f16() special cases
b3b8141
unverified
ggml : make more compatible with c99 (#262)
52bc68d
unverified
ggml : fix indentation
09b6ef7
unverified
ggml : make compatible with c99 (#262)
d9c1974
unverified
Remove C++20 requirement (#257)
6702756
unverified
Roland Rabien
commited on
talk : talk with AI in the terminal
7aad96d
ggml : add alternative cblas_sgemm call
2f68de6
unverified
fix compilation on haiku
5d3218a
Al Hoang
commited on
ggml : fix typo in previous commit
d7b9296
unverified
ggml : use macros to inline FP16 <-> FP32 conversions
23e5614
unverified
ggml : add F16C CPU flag check
2661c19
unverified
add fp16/fp32 convert intrinsics
03a5fa2
katsu560
commited on