ggerganov/llama.cpp

Pull Requests Commits
llama : fix Qwen model type strings
ggerganov committed 1 year ago
Verified f018acba
metal : hide debug messages from normal log
ggerganov committed 1 year ago
Verified 46323fa9
ggml: fix zero division in ‘dne’ calculation in CUDA COUNT_EQUAL operator when ‘ne’ is small (#10213)
SongXiaoXi committed 1 year ago
Verified 5b359bb1
ggml : optimize llamafile cpu matrix multiplication for ppc64le (#10156)
amritahs-ibm committed 1 year ago
Verified e8921349
scripts : fix pattern and get n_tokens in one go (#10221)
lhpqaq committed 1 year ago
Verified 8fc393f2
metal : opt-in compile flag for BF16 (#10218)
ggerganov committed 1 year ago
Verified ec450d3b
metal : improve clarity (minor) (#10171)
ggerganov committed 1 year ago
Verified 695ad752
metal : optimize FA kernels (#10171)
ggerganov committed 1 year ago
Verified 841f27ab
swift : exclude ggml-metal-embed.metal (#10211)
jhen0409 committed 1 year ago
Verified d05b3127
server : minor UI fix (#10207)
ngxson committed 1 year ago
Verified 76c6e7f1
server : revamp chat UI with vuejs and daisyui (#10175)
ngxson committed 1 year ago
Verified a71d81cf
scripts : add amx to sync-ggml.sh [no ci]
ggerganov committed 1 year ago
Verified eec4d717
sync : ggml
ggerganov committed 1 year ago
Verified 3b088286
scripts : sync update
ggerganov committed 1 year ago
Verified a2c6fd74
ggml : add ggml-cpu.h to the public headers (#10204)
slaren committed 1 year ago
Verified 97404c4a
Remove identical wte/etw logic for jais (#10203)
fmz committed 1 year ago
Verified 60e17ce2
DRY: Fixes clone functionality (#10192)
wwoodsTM committed 1 year ago
Verified 5107e8ce
fix q4_0_8_8 format for corrupted tokens issue (#10198)
snadampal committed 1 year ago
Verified 2319126a
Optimize RWKV6 Operator Naming and Implement Multi-core CPU/ SYCL Acceleration (#10133)
Zhiyuan Li committed 1 year ago
Verified 3bcd40b3
metal : add BF16 support (#8439)
ggerganov committed 1 year ago
Verified 5c333e01
server : remove hack for extra parallel slot (#10187)
ggerganov committed 1 year ago
Verified b11f9ba9
metal : fix from ptr buffer name (#10189)
slaren committed 1 year ago
Verified 94d8cb8b
ggml : adjust is_first_call init value (#10193)
ggerganov committed 1 year ago
Verified 1dc04b2d
metal : add quantized FA support (#10149)
ggerganov committed 1 year ago
Verified a1eaf6a9
llama : add <|tool_call|> formatting to Granite template (#10177)
gabe-l-hart committed 1 year ago
Verified b8deef0e
ggml : fix arch check in bf16_to_fp32 (#10164)
slaren committed 1 year ago
Verified a9e8a9a0
Q6_K AVX improvements (#10118)
netrunnereve committed 1 year ago
Verified 34073647
ggml : fix gelu tables initialization (#10172)
slaren committed 1 year ago
Verified d5a409e5
ggml : fix q4xx mat mul, increase ggml_aligned_malloc alignment (#10167)
slaren committed 1 year ago
Verified 401558b7
server : clarify /slots endpoint, add is_processing (#10162)
ngxson committed 1 year ago
Verified 9e0ecfb6
Terms Of Use Privacy Policy Refund Policy Impressum