ggerganov/llama.cpp

Pull Requests Commits
ci : add Ubuntu 22 Vulkan CI run (#5789)
netrunnereve committed 1 year ago
Verified 6ea0f010
server : fix newlines in help (#5785)
ggerganov committed 1 year ago
Verified f105471e
[SYCL] Use batched mul_mat pathway (#5591)
AidanBeltonS committed 1 year ago
Verified 38d15216
Server: normalize naming (#5779)
ngxson committed 1 year ago
Verified 052051d8
llama : constified `llama_set_state_data`'s `src` (#5774)
MarcusDunn committed 1 year ago
Verified d5ab2975
ci : reduce 3b ppl chunks to 1 to avoid timeout (#5771)
ggerganov committed 1 year ago
Verified 87c91c07
make portability_enumeration_ext apple only (#5757)
netrunnereve committed 1 year ago
Verified 317709b2
llama : remove deprecated API (#5770)
ggerganov committed 1 year ago
Verified 08c5ee87
awq-py : remove (#5768)
ggerganov committed 1 year ago
Verified 78aacf36
sync : ggml
ggerganov committed 1 year ago
Verified 8c0e8f4e
add google magika inference example (ggml/748)
slaren committed 1 year ago
Verified 2774b0c9
Introduce backend GUIDs (ggml/743)
uextm committed 1 year ago
Verified 5f706718
server : hit Ctrl+C twice to exit (#5734)
ngxson committed 1 year ago
Verified a693bea1
llama : fix non-quantization of expert gating tensors (#5754)
compilade committed 1 year ago
Verified adcb12a9
llama : improve BERT tokenization (#5740)
iamlemec committed 1 year ago
Verified 177628bf
readme : add link to LLaVA 1.6 models (#5758)
danbev committed 1 year ago
Verified 6c441686
server : add "/chat/completions" alias for "/v1/...` (#5722)
jorgealias committed 1 year ago
Verified efc72253
ggml : make i-quants work with super-blocks of 64 (CPU,Metal) (#5760)
ikawrakow committed 1 year ago
Verified 7c4263d4
Attempt to fix android build (#5752)
ikawrakow committed 1 year ago
Verified cb49e0f8
IQ4_XS: a 4.25 bpw quantization (#5747)
ikawrakow committed 1 year ago
Verified 0becb22a
cuda : replace remaining shfl_xor with calls to warp_reduce functions (#5744)
Engininja2 committed 1 year ago
Verified c24a2a6e
ggml-quants : fix avx2 iq1_s vec_dot when compiled with gcc (#5742)
Engininja2 committed 1 year ago
Verified 1f30b7a9
llama : fix defrag bugs + add parameter (#5735)
ggerganov committed 1 year ago
Verified 9d533a77
Makefile: use variables for cublas (#5689)
lindeer committed 1 year ago
Verified cbbd1efa
fix server hangs on empty prompt (#5733)
ngxson committed 1 year ago
Verified b11a93df
Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range (#5721)
ikawrakow committed 1 year ago
Verified a33e6a0d
CUDA: fix DEBUG_CUDA_MALLOC (#5729)
JohannesGaessler committed 1 year ago
Verified 47bb7b48
readme : update ui list (#5731)
guinmoon committed 1 year ago
Verified c4d7f817
[SYCL] Add support for soft_max ALiBi (#5639)
AidanBeltonS committed 1 year ago
Verified e849078c
unicode : reuse iterator (#5726)
ggerganov committed 1 year ago
Verified 67fd3313
Terms Of Use Privacy Policy Refund Policy Impressum