issues
search
google
/
gemma.cpp
lightweight, standalone C++ inference engine for Google's Gemma models.
Apache License 2.0
5.96k
stars
506
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Adds simple-loop versions of missing batched functions.
#226
copybara-service[bot]
closed
4 months ago
0
Update benchmark with internal init
#225
copybara-service[bot]
closed
4 months ago
0
Use CompressedWeights<TConfig<float>> in backpropagation.
#224
szabadka
closed
4 months ago
0
Fix for transpose matrix creation and additional tests
#223
copybara-service[bot]
closed
4 months ago
0
Add CPU output, error if not C++17, simplify tokenizer ctor
#222
copybara-service[bot]
closed
4 months ago
0
OFF Topic, Request for Open-Sourcing Google Gemini Flash
#221
0wwafa
closed
3 months ago
32
Shifting large matrix init to heap in ops_test.cc
#220
copybara-service[bot]
closed
4 months ago
0
Support all weight types in a single binary.
#219
copybara-service[bot]
closed
4 months ago
0
Small code cleanup suggestions while reading the code.
#218
copybara-service[bot]
closed
4 months ago
0
Add support for custom sampling function to runtime config.
#217
szabadka
closed
4 months ago
0
Fix fix for weight type define, refs #198
#216
copybara-service[bot]
closed
4 months ago
0
Fix reference to GEMMA_WEIGHT_T. Refs #198
#215
copybara-service[bot]
closed
4 months ago
0
Toward only using compressed weights:
#214
copybara-service[bot]
closed
4 months ago
0
Fix Softmax on SVE
#213
copybara-service[bot]
closed
4 months ago
0
Add Adam optimizer.
#212
szabadka
closed
4 months ago
0
Internal experiment
#211
copybara-service[bot]
closed
4 months ago
0
Implement mixed mode matmul: f32 * bf16
#210
copybara-service[bot]
closed
4 months ago
1
Simplifications: remove GemmaInterface and GemmaImpl
#209
copybara-service[bot]
closed
4 months ago
0
Remove no longer required stats.h - use Highway version instead
#208
copybara-service[bot]
closed
4 months ago
0
revert back to HWY_ASSERT for lane constraints, qualify hn::Add
#207
copybara-service[bot]
closed
4 months ago
0
Fix for GenerateZeroMat call in TestTiledMatMul
#206
copybara-service[bot]
closed
4 months ago
0
Add bf16 matmul support, update naming+test
#205
copybara-service[bot]
closed
4 months ago
0
Use system topology to pin threads across clusters.
#204
copybara-service[bot]
closed
4 months ago
0
Add first version of backpropagation support.
#203
szabadka
closed
4 months ago
1
Refactor GemmaImpl dispatch to use Highway 1.2's HWY_DYNAMIC_DISPATCH_T
#202
copybara-service[bot]
closed
4 months ago
0
Update to Highway 1.2 for topology/VQSelect
#201
copybara-service[bot]
closed
4 months ago
0
static_assert shape constraints in MatMul 4x4
#200
copybara-service[bot]
closed
4 months ago
0
Unrolled / tiled 4x4 MatMul
#199
copybara-service[bot]
closed
4 months ago
0
Gemma.cpp hangs on a Gemma 7B model that was finetuned using huggingface peft(QLoRA)
#198
webbigdata-jp
opened
4 months ago
13
Compilation fails for raspberry pi
#197
EphemeralSapient
closed
4 months ago
2
gemma.cc:1322: Failed to load model weight
#196
ordentid
closed
4 months ago
6
Generic MHA/MQA/GQA implementation
#195
copybara-service[bot]
closed
4 months ago
0
Fix normalization in Softmax function.
#194
szabadka
closed
4 months ago
0
Compiling under mingw with clang error..
#193
0wwafa
closed
3 months ago
6
Documenting the RoPE implementation.
#192
copybara-service[bot]
closed
4 months ago
0
Minor internal refactoring.
#191
copybara-service[bot]
closed
4 months ago
0
Add MMLU eval to github
#190
copybara-service[bot]
closed
4 months ago
0
Adds Kaggle testing to CI workflow
#189
pculliton
closed
4 months ago
0
Make BlobWriter::Add() accept const void*
#188
copybara-service[bot]
closed
5 months ago
0
Refer to --weights rather than --compressed_weights to simplify CLI docs
#187
copybara-service[bot]
closed
5 months ago
0
Add TTFT to TimingInfo
#186
copybara-service[bot]
closed
5 months ago
0
Paligemma Support
#185
okpatil4u
closed
1 week ago
3
Pass most runtime parameters using const RuntimeConfig&
#184
copybara-service[bot]
closed
5 months ago
0
Store tokens/sec in auxiliary struct TimingInfo.
#183
copybara-service[bot]
closed
5 months ago
1
Fix SVE build: add missing hn::
#182
copybara-service[bot]
closed
5 months ago
0
Support additional scaling
#181
copybara-service[bot]
closed
5 months ago
0
Enable even/odd for SFP. Refs #166
#180
copybara-service[bot]
closed
5 months ago
0
Fix RecurrentGemma (refs #166) - one Dot was ignoring scale.
#179
copybara-service[bot]
closed
5 months ago
0
2x speedup of SFP decode (1.4x overall) on AVX3_DL+.
#178
copybara-service[bot]
closed
5 months ago
0
Use more parallelism in attention block in prefill mode.
#177
szabadka
closed
5 months ago
0
Previous
Next