]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2023-12-24 slarensimplify bug issue template (#4623)
2023-12-24 Shintarou Okadallama : add PLaMo model (#3557)
2023-12-24 slarencuda : improve cuda pool efficiency using virtual memor...
2023-12-23 slarenfallback to CPU buffer if host buffer alloc fails ...
2023-12-23 Samuel Maynardci(docker): fix tags in "Build and push docker image...
2023-12-23 Alexey Parfenovserver : allow to specify custom prompt for penalty...
2023-12-23 kalomazegrammar : check the full vocab only if necessary (opt...
2023-12-23 Johannes GäßlerCUDA: fixed row rounding for 0 tensor splits (#4594)
2023-12-22 LeonEricssonlookup : add prompt lookup decoding example (#4484)
2023-12-22 Georgi Gerganovsync : ggml (fix im2col) (#4591)
2023-12-22 FantasyGmmcuda : fix jetson compile error (#4560)
2023-12-22 Henrik ForsténFix CudaMemcpy direction (#4599)
2023-12-22 slarenllama : fix platforms without mmap (#4578)
2023-12-22 Herman Semenovggml : add comment about backward GGML_OP_DIAG_MASK_INF...
2023-12-22 Michael Kespermake : add LLAMA_HIP_UMA option (#4587)
2023-12-22 rhuddlestonci : tag docker image with build number (#4584)
2023-12-22 Deinsreadme : add zig bindings (#4581)
2023-12-22 bobqianicggml : extend `enum ggml_log_level` with `GGML_LOG_LEVE...
2023-12-22 crasmllama : add ability to cancel model loading (#4462)
2023-12-21 Georgi Gerganovggml : change ggml_scale to take a float instead of...
2023-12-21 Georgi Gerganovgguf-py : fix broken link
2023-12-21 Georgi Gerganovgguf : simplify example dependencies
2023-12-21 Samuel Maynardci : add `jlumbroso/free-disk-space` to docker workflow...
2023-12-21 slarenllama : initial ggml-backend integration (#4520)
2023-12-21 Marcus Dunnllama : allow getting n_batch from llama_context in...
2023-12-21 Finn Voorheesmetal : fix `ggml_metal_log` vargs (#4373)
2023-12-21 Erik Garrisoncuda : ROCm AMD Unified Memory Architecture (UMA) handl...
2023-12-21 arlo-phoenixggml-cuda: Fix HIP build by adding define for __trap...
2023-12-21 Jared Van Bortelcommon : remove incorrect --model-draft default (#4568)
2023-12-21 Johannes GäßlerCUDA: mul_mat_id always on GPU for batches >= 32 (...
2023-12-21 Georgi Gerganovreadme : update coding guidelines
2023-12-21 howlgerpy : open merges file as 'utf-8' (#4566)
2023-12-21 bobqianiccuda : better error message for ggml_get_rows (#4561)
2023-12-21 slarencuda : replace asserts in wrong architecture checks...
2023-12-21 Johannes Gäßlerllama : disable per-tensor info prints on model load...
2023-12-21 LoganDarkFix access violation in ggml_cuda_free_data if tensor...
2023-12-20 Johannes GäßlerCUDA: Faster Mixtral prompt processing (#4538)
2023-12-19 Eric Sommerladeggml : fixed check for _MSC_VER (#4535)
2023-12-18 arlo-phoenixggml-cuda: Fix HIP build (#4528)
2023-12-18 Georgi Gerganovllama.swiftui : add tinyllama 1.1B F16
2023-12-18 Georgi Gerganovllama.swiftui : add more models
2023-12-18 Ebey Abrahamllama : add phi-2 + fix NeoX rope + ggml_mul_mat_set_pr...
2023-12-18 hankcsllama : fix try_override for bool_value which always...
2023-12-18 Jared Van Borteldecode : fix logits_valid for legacy API (#4516)
2023-12-17 Georgi Gerganovreadme : update hot topics
2023-12-17 Georgi Gerganovllama.swiftui : add bench functionality (#4483)
2023-12-17 Jared Van Bortelgguf-py : fail fast on nonsensical special token IDs...
2023-12-17 Matheus Gabriel... build : Check the ROCm installation location (#4485)
2023-12-17 slarenfinetune : keep allocs alive until all allocations...
2023-12-17 olexiybserver : disable llm logs if SERVER_VERBOSE is off...
2023-12-17 AdithyanIserver : fix grammar being ignored (#4494)
2023-12-17 Alexey Parfenovserver : fix possible ambiguity in content type charset...
2023-12-17 mzcuserver : allow requests larger than 8K (#4500)
2023-12-17 Bach LeLink to cublas dynamically on Windows even with LLAMA_S...
2023-12-16 slarenlora : add support for non-llama models (#3333)
2023-12-16 Jared Van Bortelllama : sanity checks for access to logits (#4274)
2023-12-15 ShadovvBeastserver : add optional API Key Authentication example...
2023-12-15 slarenggml : group mul_mat_id rows by matrix (cpu only) ...
2023-12-14 slarenggml : use ggml_row_size where possible (#4472)
2023-12-14 slarenggml : remove n_dims from ggml_tensor (#4469)
2023-12-14 wonjun Jangpy : add protobuf dependency (#4466)
2023-12-14 LostRuinsggml : add ggml_row_size() (fixes llama out of space...
2023-12-14 Georgi Gerganovggml : fix OpenCL broadcast requirement for ggml_mul...
2023-12-14 wonjun Jangconvert : support loading vocab from fast tokenizer...
2023-12-14 BarfingLemursreadme : update supported model list (#4457)
2023-12-13 shibe2server : fix handling of characters that span multiple...
2023-12-13 Georgi Gerganovsync : ggml (SD ops, tests, kernels) (#4444)
2023-12-13 Jared Van Bortelbuild : detect host compiler and cuda compiler separate...
2023-12-13 Siwen Yucommon : add `--version` option to show build info...
2023-12-13 Georgi Gerganovreadme : update hot topics
2023-12-13 slarenllama : add Mixtral support (#4406)
2023-12-12 kalomazeserver : tweak default sampling parameters (#4367)
2023-12-12 Richard Kissenglish : use `typos` to fix comments and logs (#4354)
2023-12-12 Jared Van Bortelbuild : target Windows 8 for standard mingw-w64 (#4405)
2023-12-12 crasmllama : document logits_all deprecation (#4418)
2023-12-12 Vladimir Zorinserver : fix local model name in server (#4420)
2023-12-12 Taikono-Himazinggml : increased GGML_MAX_PARAMS to allow finetuning...
2023-12-10 Yueh-Po PengUpdate README.md (#4388)
2023-12-09 Xiang (Kevin) Ligrammar : revert the replacement of llama_token_to_piec...
2023-12-07 Georgi Gerganovsync : ggml (new ops, tests, backend, etc.) (#4359)
2023-12-07 Georgi Gerganovllama : per-layer KV cache + quantum K cache (#4309)
2023-12-07 Hongyu Ouyangtrain : fix #4227 (double free in examples/train-text...
2023-12-06 Georgi Gerganovserver : recognize cache_prompt parameter in OAI API...
2023-12-06 Georgi Gerganovcommon : fix compile warning
2023-12-06 stduhpfspeculative : support `--color` (#4343)
2023-12-05 Marcus Dunngrammar : pre-computed pieces + reserve mem + less...
2023-12-05 Kerfufflellama : allow overriding GGUF metadata when loading...
2023-12-05 MaggotHATEsampling : custom samplers order (#4285)
2023-12-05 kchro3swift : revert compiler checks for swift package (...
2023-12-04 Daniel Beveniussimple : update error message for KV cache check (...
2023-12-04 Miwa / Ensanswift : fix concatenation method to avoid invalid UTF8...
2023-12-04 Miwa / Ensanswift : fix prompt tokenization logic (#4321)
2023-12-04 Ikko Eltociear... grammar-parser : fix typo (#4318)
2023-12-03 Georgi Gerganovggml : reuse ggml_get_n_tasks() in ggml_graph_plan...
2023-12-03 Georgi Gerganovggml : fix soft max out-of-bounds access (#4307)
2023-12-03 Ed Leeserver : fix OpenAI API `stop` field to be optional...
2023-12-03 Rickard Edénpy : add grammar to oai like api (#4294)
2023-12-03 Georgi Gerganovllama : pad KV cache size (#4280)
2023-12-01 Georgi Gerganovllama : avoid using "optional" keyword (#4283)
2023-12-01 Georgi Gerganovllama : support optional tensors (#4283)
next