Commit History

Autor SHA1 Mensaxe Data
  Daniel Bevenius 8a56075b07 gritlm : add --outdir option to hf.sh script (#6699) hai 1 ano
  Georgi Gerganov 58227ffdeb perplexity : require positive --ctx-size arg (#6695) hai 1 ano
  Daniel Bevenius 4fbd8098e6 gguf : add special tokens metadata for FIM/Infill (#6689) hai 1 ano
  Olivier Chafik 7593639ce3 `main`: add --json-schema / -j flag (#6659) hai 1 ano
  compilade 132f55795e llama : fix restoring the number of outputs from state files (#6687) hai 1 ano
  Pierrick Hymbert 3272896d79 server : revert "minor layout improvements" (#6684) hai 1 ano
  Steven Prichard 7fc16a2c32 swift : linux support (#6590) hai 1 ano
  Neo Zhang Jianyu 17e98d4c96 fix mul_mat_id() for new input, make the ut pass (#6682) hai 1 ano
  David Renshaw 1958f7e06c llama : add missing kv clear in llama_beam_search (#6664) hai 1 ano
  Chao Jiang 04fbc5f23e Add Command R chat template (#6650) hai 1 ano
  Georgi Gerganov f184dd9208 flake.lock: Update (#6669) hai 1 ano
  Dave 422c2aff1c Added support for GGML_OP_CLAMP in Metal (#6662) hai 1 ano
  Sigbjørn Skjæret 8800226d65 Fix --split-max-size (#6655) hai 1 ano
  Jaemin Son e689fc4e91 [bug fix] convert github repository_owner to lowercase (#6673) hai 1 ano
  James A Capozzoli a4ec34e1cd convert : enable the `--use-temp-file` cli flag (#6645) hai 1 ano
  Neo Zhang Jianyu de17e3f745 fix memcpy() crash, add missed cmd in guide, fix softmax (#6622) hai 1 ano
  Johannes Gäßler b5e7285baf CUDA: fix matrix multiplication logic for tests (#6667) hai 1 ano
  Pierrick Hymbert 4bd0f93e4a model: support arch `DbrxForCausalLM` (#6515) hai 1 ano
  Olivier Chafik ab9a3240a9 JSON schema conversion: ⚡️ faster repetitions, min/maxLength for strings, cap number length (#6555) hai 1 ano
  slaren fbbc030ba9 metal : unify mul_mv_id kernels (#6556) hai 1 ano
  Daniel Bevenius 4cc120c744 infill : add download instructions for model (#6626) hai 1 ano
  Pierrick Hymbert 24ee66ed0d server : coherent log output for KV cache full (#6637) hai 1 ano
  jiez 91c736015b llama : add gguf_remove_key + remove split meta during quantize (#6591) hai 1 ano
  Rene Leonhardt 5c4d767ac0 chore: Fix markdown warnings (#6625) hai 1 ano
  Georgi Gerganov ef21ce4ccb imatrix : remove invalid assert (#6632) hai 1 ano
  MasterYi1024 dee7f8d692 Correct free memory and total memory. (#6630) hai 1 ano
  Pierrick Hymbert 81da18e71c eval-callback: use ggml_op_desc to pretty print unary operator name (#6631) hai 1 ano
  Georgi Gerganov 9ed2737acc ci : disable Metal for macOS-latest-cmake-x64 (#6628) hai 1 ano
  Clint Herron 04a5ac211e Optimization: eliminate addition of redundant stacks when advancing grammar. (#6616) hai 1 ano
  Clint Herron f7001ccc5a As suggested by @slaren, disabling Metal for test to fix CI build on OSX from #6576 (#6619) hai 1 ano