bench-all : add distil models

bench-all : indentations

whisper : make large version explicit + fix data size units (#1493)

whisper : add batched decoding (#1486)

whisper : add full CUDA and Metal offloading (#1472)

whisper : add support for large v3 (#1444)

cuda : fix HIPBLAS build

sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)

extra: Add benchmark script implemented in Python (#1298)

metal : add F32 support + update bench output

whisper : Metal and ggml-alloc support (#1270)

extra : update 'quantize-all.sh' to quantize all downloaded models (#1054)

metal : sync ggml-metal (ref #1047)

opencl : sync latest ggml-opencl

extra : update ggml sync script

ggml : sync ggml (clBLAST + tensor names)

bench : improve benchmarks

whisper : add integer quantization support (#540)

extra : add sync-ggml.sh script

whisper : add Core ML support (#566)

bench-wts.sh : rename script + add execute permission

qual-bench.sh : add quality comparison tool, and update main.cpp to allow using a font file (#569)

bench : minor fixes

bench : add memcpy and ggml_mul_mat benchmarks

Node.js package (#260)

bench : more concise representation of the results (#89)

bench.wasm : same as "bench" but runs in the browser (#89)

models : add the new "large" model release by OpenAI

bench : add commit hash to bench-all.sh results

command.wasm : add voice assistant example for the Web (#171)

extra : compute SHA of all models files

Add helper script to benchmark all models

Update README.md and finalize the whisper.wasm example

Adding helper script for converting the PT models