llama.cpp
Inference support for T5 and FLAN-T5 model families
#8141
Merged
Overview
Commits
18
Changes
View On GitHub
Commits
  • llama : add inference support and model types for T5 and FLAN-T5 model families
    sszymczy committed 1 year ago
  • Merge branch 'ggerganov:master' into t5-clean-3
    fairydreaming committed 1 year ago
  • llama : updated llm_build_ffn() calls to new API in build_t5()
    sszymczy committed 1 year ago
  • llama : make pos_bias contiguous for CUDA
    sszymczy committed 1 year ago
  • Merge remote-tracking branch 'upstream/master' into t5-clean-3
    sszymczy committed 1 year ago
  • llama : whitespace formatting
    sszymczy committed 1 year ago
  • llama : quantization-related fixes for T5
    sszymczy committed 1 year ago
  • llama : add early return in Unigram tokenizer when normalized input is empty
    sszymczy committed 1 year ago
  • llama : remove obsolete code
    sszymczy committed 1 year ago
  • add t5 tokenizer tests
    ggerganov committed 1 year ago
  • Merge remote-tracking branch 'upstream/master' into t5-clean-3
    sszymczy committed 1 year ago
  • llama : move JAIS after T5 everywhere for easier merging later
    sszymczy committed 1 year ago
  • llama : change naming to prefer "_enc" suffix
    ggerganov committed 1 year ago
  • llama : simplify llama_encode_internal
    ggerganov committed 1 year ago
  • llama-batched : add encoder support
    ggerganov committed 1 year ago
  • llama : minor
    ggerganov committed 1 year ago
  • llama : silence compiler warnings
    sszymczy committed 1 year ago
  • Merge branch 'ggerganov:master' into t5-clean-3
    fairydreaming committed 1 year ago
FAQ Terms Privacy Refunds Impressum
Loading