]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
mtmd : add qwen2vl and qwen2.5vl (#13141)
authorXuan-Son Nguyen <redacted>
Tue, 29 Apr 2025 09:47:04 +0000 (11:47 +0200)
committerGitHub <redacted>
Tue, 29 Apr 2025 09:47:04 +0000 (11:47 +0200)
commit00e3e5a194e88e604e7c91391b9e90332888fd72
treec51e7a873a8c84f7c8bf74ca913ab2b90dcb1fc0
parente98b3692be4cd8fbbd9a56fbacc2f2bf0bf26a68
mtmd : add qwen2vl and qwen2.5vl (#13141)

* llava : add clip_n_output_tokens, deprecate clip_n_patches

* mtmd : add qwen2vl and qwen2.5vl

* decode_embd_batch::set_position_...

* working version

* deprecate llama-qwen2vl-cli

* correct order W, H of clip_embd_nbytes_by_img

* edit existing line in hot topics
README.md
examples/llava/CMakeLists.txt
examples/llava/clip.cpp
examples/llava/clip.h
examples/llava/llava.cpp
examples/llava/mtmd-cli.cpp
examples/llava/mtmd.cpp
examples/llava/mtmd.h
examples/llava/qwen2vl-cli.cpp [deleted file]
examples/llava/qwen2vl-test.cpp [new file with mode: 0644]
examples/llava/tests.sh