8000 Commits · prajwalc22/gemma.cpp · GitHub

More Web Proxy on the site http://driver.im/

Commits

Commits on Mar 25, 2025

Merge pull request #527 from ufownl:feature/gemma2_secondary_eos
copybara-github
committed

Commits on Mar 21, 2025

Set the secondary EOS for Gemma2
ufownl
committed
Fix the EOS checking
ufownl
committed
Update app argument documentation

jan-wassenberg
authored and
copybara-github
committed

Commits on Mar 20, 2025

Add support for a secondary EOS token

pculliton
authored and
copybara-github
committed
Add note on attention length and SFP

jan-wassenberg
authored and
copybara-github
committed

Commits on Mar 19, 2025

Merge pull request #523 from ufownl/bugfix/gemma3_1b_wrapping
pculliton
authored

Commits on Mar 18, 2025

Fix the prompt wrapping of gemma3-1b
ufownl
committed

Commits on Mar 14, 2025

Refactor Gemma ctor and improve pool NUMA support

jan-wassenberg
authored and
copybara-github
committed

Commits on Mar 13, 2025

Fix PaliGemma models.

pculliton
authored and
copybara-github
committed

Commits on Mar 12, 2025

Point out Gemma 3 support in README.md

Quirin Niedernhuber
authored and
copybara-github
committed
Update github actions/cache version

jan-wassenberg
authored and
copybara-github
committed
Internal change.

pculliton
authored and
copybara-github
committed
Internal change.

pculliton
authored and
copybara-github
committed

Commits on Feb 26, 2025

Support bf16 output of Matmul

jan-wassenberg
authored and
copybara-github
committed

Commits on Feb 24, 2025

With new matmul, much larger batch sizes are advantageous, default to 256.

jan-wassenberg
authored and
copybara-github
committed

Commits on Feb 21, 2025

Move batch_bench into test section, add GTest dep. Fixes #501

jan-wassenberg
authored and
copybara-github
committed

Commits on Feb 20, 2025

Matmul rewrite: fp64 sums, hierarchical parallelization, cache-blocking, autotuning

jan-wassenberg
authored and
copybara-github
committed

Commits on Feb 18, 2025

Use vectorized TopK using highway VQSelect

apoorvreddy
authored and
copybara-github
committed

Commits on Feb 17, 2025

Implements FusedSoftmaxAndSampleTopK.

apoorvreddy
authored and
copybara-github
committed

Commits on Feb 15, 2025

Only temporarily enable spinning in threading benchmark

jan-wassenberg
authored and
copybara-github
committed

Commits on Feb 13, 2025

Less verbose threading_test output, improve formatting.

jan-wassenberg
authored and
copybara-github
committed

Commits on Feb 12, 2025

Add conversion tool for HF safetensors to gemma.cpp for PaliGemma.

danielkeysers
authored and
copybara-github
committed

Commits on Feb 11, 2025

Commits on Feb 9, 2025

Further speed up blob_compare: single alloc, use dual sockets

jan-wassenberg
authored and
copybara-github
committed

Commits on Feb 8, 2025

Fix the link error when building `compress_weights` with Clang on macOS
ufownl
committed

Commits on Feb 7, 2025

Commits on Feb 6, 2025

Improved blob diff: parallel, tolerance for float

jan-wassenberg
authored and
copybara-github
committed

Commits on Feb 5, 2025

Public visibility for compression/

jan-wassenberg
authored and
copybara-github
committed

0