ggml : try fix ppc64 (#0)

ggml : optimize for ppc64le using VSX intrinsics (ggml/784)

ggml : introduce bfloat16 support (llama/6412)

add basic tensor data validation function (llama/6884)

ggml : fix MIN / MAX macros (llama/6904)

ggml : move 32-bit arm compat in ggml-impl.h (llama/6865)

ggml : add llamafile sgemm (llama/6414)

build: fix and ignore msvc warnings (ggml/805)

llama : add Command R Plus support (llama/6491)

sync : ggml (#2001)

ggml : reuse quantum structs across backends (llama/5943)

ggml : fix UB in IQ2_S and IQ3_S (llama/6012)

1.5 bit: we can do even better (llama/5999)

ggml, ci : Windows ARM runner and build fixes (llama/5979)

Better 1.5 bit quantization (llama/5971)

ggml : fix unnecessary f32 -> f16 -> f32 casts (mmla) (llama/5951)

ggml : remove old quantization functions (llama/5942)

ggml : add ggml-common.h to deduplicate shared code (llama/5940)

ggml : try fix 32-bit arm compat (#1938)

ggml : use `uint8x16_t` return type for `ggml_vqtbl1q_u8` (llama/5894)

quants : use MM256_SET_M128I consistently to fix gcc 7 build (llama/5889)

ggml : fix IQ3_S AVX implementation (llama/5834)

ggml : IQ3_S improvements (llama/5829)

ggml : make i-quants work with super-blocks of 64 (CPU,Metal) (llama/5760)

Attempt to fix android build (llama/5752)

IQ4_XS: a 4.25 bpw quantization (llama/5747)

ggml-quants : fix avx2 iq1_s vec_dot when compiled with gcc (llama/5742)

Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range (llama/5721)

ggml-quants : provide ggml_vqtbl1q_u8 for 64bit compatibility (llama/5711)

IQ3_S: a much better alternative to Q3_K (llama/5676)

ggml : always define ggml_fp16_t as uint16_t (llama/5666)

ggml : 32-bit arm compat (#1891)

sync : llama.cpp (ggml/0)

ggml : restore vec dot stride arg names (llama/5453)

ci : fix wikitext url + compile warnings (llama/5569)

1.5 bit quantization (llama/5453)

ggml-quants : fix compiler warnings (shadow variable) (llama/5472)

ggml : fix compile warnings (unused vars) (llama/4966)

ggml : add mmla kernels for quantized GEMM (llama/4966)

ggml : fix `error C2078: too many initializers` for MSVC ARM64 (llama/5404)

Slight quantization improvement for Q4_K and Q5_K (llama/5361)

iq2_xxs: tune quantization (llama/5320)

Faster AVX2 dot product for IQ2_XS (llama/5187)

SOTA 3-bit quants (llama/5196)

ggml : add IQ2 to test-backend-ops + refactoring (llama/4990)

ggml : importance matrix support for legacy quants (llama/4969)

Add ability to use importance matrix for all k-quants (llama/4930)

2-bit quantizations (llama/4897)

ggml : fix 32-bit ARM compat for IQ2_XS (#1758)

ggml : SOTA 2-bit quants (add IQ2_XS) (llama/4856)