0d56246f4b
* ggml : group all experts in a single ggml_mul_mat_id cuda : improve mmid row copy * cuda : fix bin bcast with non-cont src0 * test-backend-ops : only run all mul mat tests for base types * llama : disable moe offloading with SYCL --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> |
||
|---|---|---|
| .. | ||
| build-info.cmake | ||
| build-info.sh | ||
| check-requirements.sh | ||
| ci-run.sh | ||
| compare-commits.sh | ||
| compare-llama-bench.py | ||
| convert-gg.sh | ||
| gen-authors.sh | ||
| gen-build-info-cpp.cmake | ||
| get-flags.mk | ||
| get-hellaswag.sh | ||
| get-pg.sh | ||
| get-wikitext-2.sh | ||
| get-wikitext-103.sh | ||
| get-winogrande.sh | ||
| hf.sh | ||
| install-oneapi.bat | ||
| LlamaConfig.cmake.in | ||
| pod-llama.sh | ||
| qnt-all.sh | ||
| run-all-perf.sh | ||
| run-all-ppl.sh | ||
| run-with-preset.py | ||
| server-llm.sh | ||
| sync-ggml-am.sh | ||
| sync-ggml.last | ||
| sync-ggml.sh | ||
| verify-checksum-models.py | ||