tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-08 00:52:56 +00:00
Code Issues Packages Projects Releases Wiki Activity

cuda : implement ssm scan for Mamba2

Browse Source 
There is still room for improvement, but it works!

* cuda : adapt Mamba1 ssm scan to shape changes from Mamba2
This commit is contained in:
Francis Couture-Harpin
2025-05-15 18:09:53 -04:00
parent a42f239418
commit f8c7caeeb7
4 changed files with 195 additions and 54 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/test-backend-ops.cpp
View File

@ -4225,7 +4225,7 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
test_cases.emplace_back(new test_ssm_conv(GGML_TYPE_F32, {4, 1536, 4, 1}, {4, 1536, 1, 1}));
test_cases.emplace_back(new test_ssm_scan(GGML_TYPE_F32, 16, 1, 1024, 1, 32, 4)); // Mamba-1
test_cases.emplace_back(new test_ssm_scan(GGML_TYPE_F32, 128, 32, 32, 2, 32, 4)); // Mamba-2
test_cases.emplace_back(new test_ssm_scan(GGML_TYPE_F32, 128, 64, 16, 2, 32, 4)); // Mamba-2
test_cases.emplace_back(new test_rwkv_wkv6(GGML_TYPE_F32, 32, 64, 1, 1));
test_cases.emplace_back(new test_rwkv_wkv6(GGML_TYPE_F32, 32, 64, 32, 1));