]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2024-11-08 Jhen-Jie Hongswift : exclude ggml-metal-embed.metal (#10211)
2024-11-07 Xuan Son Nguyenserver : minor UI fix (#10207)
2024-11-07 Xuan Son Nguyenserver : revamp chat UI with vuejs and daisyui (#10175)
2024-11-07 Georgi Gerganovscripts : add amx to sync-ggml.sh [no ci]
2024-11-07 Georgi Gerganovsync : ggml
2024-11-07 Georgi Gerganovscripts : sync update
2024-11-07 Diego Devesaggml : add ggml-cpu.h to the public headers (#10204)
2024-11-07 Faisal ZaghloulRemove identical wte/etw logic for jais (#10203)
2024-11-07 wwoodsTMDRY: Fixes clone functionality (#10192)
2024-11-07 snadampalfix q4_0_8_8 format for corrupted tokens issue (#10198)
2024-11-07 Zhiyuan LiOptimize RWKV6 Operator Naming and Implement Multi...
2024-11-06 Georgi Gerganovmetal : add BF16 support (#8439)
2024-11-06 Georgi Gerganovserver : remove hack for extra parallel slot (#10187)
2024-11-06 Diego Devesametal : fix from ptr buffer name (#10189)
2024-11-06 Georgi Gerganovggml : adjust is_first_call init value (#10193)
2024-11-06 Georgi Gerganovmetal : add quantized FA support (#10149)
2024-11-05 Gabe Goodhartllama : add <|tool_call|> formatting to Granite templat...
2024-11-04 Diego Devesaggml : fix arch check in bf16_to_fp32 (#10164)
2024-11-04 EveQ6_K AVX improvements (#10118)
2024-11-04 Diego Devesaggml : fix gelu tables initialization (#10172)
2024-11-04 Diego Devesaggml : fix q4xx mat mul, increase ggml_aligned_malloc...
2024-11-04 Xuan Son Nguyenserver : clarify /slots endpoint, add is_processing...
2024-11-04 snadampalfix build break on arm64 linux (#10166)
2024-11-04 Diego Devesacuda : clear error after changing peer access (#10153)
2024-11-04 Georgi Gerganovmetal : simplify f16 and f32 dequant kernels (#0)
2024-11-04 Georgi Gerganovmetal : move dequantize templates to beginning of MSL...
2024-11-04 leo-ponyCANN: adjust backend registry refactor. (#10158)
2024-11-04 Georgi Gerganovsync : ggml
2024-11-04 Yuri Khrustalevcmake : make it possible linking ggml as external lib...
2024-11-04 Plamen Minevmetal : fix minor string leaks (ggml/1004)
2024-11-03 Diego Devesaggml : move CPU backend to a separate file (#10144)
2024-11-03 Georgi Gerganovmetal : minor fixup in FA kernel (#10143)
2024-11-03 Georgi Gerganovflake.lock: Update (#10146)
2024-11-02 Christian KöhnenkampAdd apple arm to presets (#10134)
2024-11-02 sasha0552server : fix slot selection by lru (#10126)
2024-11-02 Georgi Gerganovserver : fix endpoint checks (#10135)
2024-11-02 Georgi Gerganovllama : adjust default context size + print warnings...
2024-11-02 Diego Devesasimple-chat : only add bos on first prompt (#10129)
2024-11-02 Xuan Son Nguyenconvert-lora : make `--base` optional (#10110)
2024-11-01 Diego Devesallama : add simple-chat example (#10124)
2024-11-01 Diego Devesallama : use smart pointers for ggml resources (#10117)
2024-11-01 Shupei Fanvulkan : improve ggml_vk_create_buffer error handling...
2024-11-01 Georgi Gerganovreadme : update hot topics
2024-11-01 sasha0552server : fix smart selection of available slot (#10120)
2024-11-01 Georgi Gerganovggml : remove ggml_scratch (#10121)
2024-11-01 Georgi Gerganovsync : ggml
2024-11-01 Georgi Gerganovggml : alloc ggml_contexts on the heap (whisper/2525)
2024-11-01 Zhenwei Jinbuild: fix build error in Windows env with OneAPI setup...
2024-10-31 Diego Devesallama : improve output buffer type selection (#10098)
2024-10-31 Diego Devesaquantize : fix --keep-split (#10114)
2024-10-31 Diego Devesallama : fix buffer checks for mamba and rwk (#10111)
2024-10-31 Zhenwei Jinloader: refactor tensor weights storage (#9935)
2024-10-31 Kevin Gibbonsserver : include scheme when printing URL (#10106)
2024-10-31 Diego Devesaggml : check tensor name lengths in gguf files (#10100)
2024-10-31 Sergio Lópezkompute: add mul_mat_q4_k shader (#10097)
2024-10-30 Sergio Lópezkompute: add backend registry / device interfaces ...
2024-10-30 Diego Devesaggml : fix memory leaks when loading invalid gguf files...
2024-10-30 Rich Doughertyreadme : more lora detail in main example readme (...
2024-10-30 Rich Doughertyconvert : more detailed convert lora usage docs (#10065)
2024-10-30 xctanggml : add Q4_0_8_8 RISC-V GEMV and GEMM kernels (...
2024-10-30 Diego Devesallama : refactor model loader with backend registry...
2024-10-29 Changyeon Kimggml: Add POOL2D OP for GPU acceleration to the Vulkan...
2024-10-29 Georgi Gerganovllama : remove Tail-Free sampling (#10071)
2024-10-28 arch-btwllama : Add IBM granite template (#10013)
2024-10-28 Georgi Gerganovflake.lock: Update (#10063)
2024-10-28 R0CKSTARmusa: workaround for Guilty Lockup in cleaning src0...
2024-10-28 Georgi Gerganovserver : don't overfill the batch during infill (#10018)
2024-10-27 Georgi Gerganovllama : switch KQ multiplication to F32 precision by...
2024-10-26 Georgi Gerganovsync : ggml
2024-10-26 bssrdfincrease cuda_cpy block size (ggml/996)
2024-10-26 Georgi Gerganovscripts : fix amx sync [no ci]
2024-10-25 Georgi Gerganovmetal : support permuted matrix multiplicaions (#10033)
2024-10-25 wwoodsTMllama : add DRY sampler (#9702)
2024-10-25 Michael Podvitskiyllama: string_split fix (#10022)
2024-10-25 Srihari-mcwllamafile : extend sgemm.cpp support for Q5_0 models...
2024-10-25 Georgi Gerganovserver : check that the prompt fits in the slot's conte...
2024-10-24 Xuan Son Nguyenserver : refactor slot input data, move tokenizer to...
2024-10-24 Georgi Gerganovci : fix cmake flags for SYCL
2024-10-24 Johannes GäßlerCUDA: fix insufficient buffer clearing for MMQ (#10032)
2024-10-24 Johannes GäßlerCUDA: fix MMQ for non-contiguous src0, add tests (...
2024-10-23 wwoodsTMserver : samplers accept the prompt correctly (#10019)
2024-10-23 Georgi Gerganovsync : ggml
2024-10-23 Georgi Gerganovllama.vim : bump generation time limit to 3s [no ci]
2024-10-23 Johannes GäßlerCUDA: fix 1D im2col, add tests (ggml/993)
2024-10-23 Daniel Beveniusggml : remove redundant set of contexts used field...
2024-10-23 Michael Coppolallama.vim : add classic vim support (#9995)
2024-10-23 Jun Hee Yoometal : add POOL2D and fix IM2COL (#9943)
2024-10-23 github-actions... flake.lock: Update
2024-10-22 Xuan Son Nguyenllama : fix empty batch causing llama_batch_allocr...
2024-10-22 Daniel Beveniusllama : rename batch to ubatch (#9950)
2024-10-22 Molly SophiaRwkv chat template fix (#10001)
2024-10-22 Xuan Son Nguyenlora : warn user if new token is added in the adapter...
2024-10-22 Molly Sophiallama : add chat template for RWKV-World + fix EOT...
2024-10-22 leo-pony[CANN] Adapt to dynamically loadable backends mechanism...
2024-10-22 Daniel Beveniusarg : fix typo in embeddings argument help [no ci]...
2024-10-21 Georgi Gerganovllama.vim : fix info text display [no ci] (#9787)
2024-10-21 Georgi Gerganovllama.vim : move info to the right of screen [no ci...
2024-10-21 Asghar Ghorbanireadme : update UI list (#9972)
2024-10-21 Daniel Beveniusarg : fix attention non-causal arg value hint (#9985)
2024-10-21 Georgi Gerganovllama.vim : plugin for Neovim (#9787)
next