[Skip to content](https://github.com/thecodacus/llama.cpp#start-of-content)
You signed in with another tab or window. [Reload](https://github.com/thecodacus/llama.cpp) to refresh your session.You signed out in another tab or window. [Reload](https://github.com/thecodacus/llama.cpp) to refresh your session.You switched accounts on another tab or window. [Reload](https://github.com/thecodacus/llama.cpp) to refresh your session.Dismiss alert
{{ message }}
[thecodacus](https://github.com/thecodacus)/ **[llama.cpp](https://github.com/thecodacus/llama.cpp)** Public
forked from [ggml-org/llama.cpp](https://github.com/ggml-org/llama.cpp)
- [Notifications](https://github.com/login?return_to=%2Fthecodacus%2Fllama.cpp) You must be signed in to change notification settings
- [Fork\\
1](https://github.com/login?return_to=%2Fthecodacus%2Fllama.cpp)
- [Star\\
1](https://github.com/login?return_to=%2Fthecodacus%2Fllama.cpp)
master
[**578** Branches](https://github.com/thecodacus/llama.cpp/branches) [**6200** Tags](https://github.com/thecodacus/llama.cpp/tags)
[Go to Branches page](https://github.com/thecodacus/llama.cpp/branches)[Go to Tags page](https://github.com/thecodacus/llama.cpp/tags)
Go to file
Code
Open more actions menu
This branch is [829 commits behind](https://github.com/thecodacus/llama.cpp/compare/master...ggml-org%3Allama.cpp%3Amaster) ggml-org/llama.cpp:master.
## Folders and files
| Name | Name | Last commit message | Last commit date |
| --- | --- | --- | --- |
| ## Latest commit
[tc-mb](https://github.com/thecodacus/llama.cpp/commits?author=tc-mb)
and
[CISC](https://github.com/thecodacus/llama.cpp/commits?author=CISC)
[mtmd : support MiniCPM-V 4.6 (](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360) [ggml-org#22529](https://github.com/ggml-org/llama.cpp/pull/22529) [)](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360)
Open commit details
2 months agoMay 6, 2026
[2496f9c](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360) · 2 months agoMay 6, 2026
## History
[9,049 Commits](https://github.com/thecodacus/llama.cpp/commits/master/)
Open commit details
[View commit history for this file.](https://github.com/thecodacus/llama.cpp/commits/master/) 9,049 Commits |
| [.devops](https://github.com/thecodacus/llama.cpp/tree/master/.devops ".devops") | [.devops](https://github.com/thecodacus/llama.cpp/tree/master/.devops ".devops") | [\[SYCL\] Update oneapi 2025.3.3, Seperate SYCL build, release Ubuntu 24…](https://github.com/thecodacus/llama.cpp/commit/4ead6fd9571df75c21599f3bafabb457e16b22c4 "[SYCL] Update oneapi 2025.3.3, Seperate SYCL build, release Ubuntu 24 package. (#22078) * upgrade oneAPI to 2025.3.3 * update * seperate SYCL CI and support release binary package for ubuntu 24 * add dependence * remove wrong copy lines * add missed line * remove other task to test the release for SYCL * rm more for test release * fix file name * correct the error in running * support build for fp32/fp16 * rm ubuntu-24-sycl-fp16 for duplicated * refactor build setting * update guide for ubuntu 24 release package, restore the release.yml for other backend * user docker replace to install oneAPI * use download installation package to replace docker * use wget to download and install oneapi, replace the apt cmd * enable ccache for oneAPI installation * fix format error * enable cache for oneAPI installation * update guide * Update .github/workflows/release.yml Co-authored-by: Sigbjørn Skjæret * Update .github/workflows/release.yml Co-authored-by: Sigbjørn Skjæret * Update .github/workflows/build-sycl.yml Co-authored-by: Sigbjørn Skjæret * Update .github/workflows/release.yml Co-authored-by: Sigbjørn Skjæret --------- Co-authored-by: Sigbjørn Skjæret ") | 3 months agoApr 23, 2026 |
| [.gemini](https://github.com/thecodacus/llama.cpp/tree/master/.gemini ".gemini") | [.gemini](https://github.com/thecodacus/llama.cpp/tree/master/.gemini ".gemini") | [contributing: tighten AI usage policy (](https://github.com/thecodacus/llama.cpp/commit/3595ae5963f1583f53beecf9725c919d309e15da "contributing: tighten AI usage policy (#18388) * contributing: tighten AI usage policy * refactor AGENTS.md * proofreading * update contributing * add claude.md * add trailing newline * add note about dishonest practices * rm point about dishonest * rm requirement watermarking * add .gemini/settings.json * allow initially AI-generated content * revise * Update CONTRIBUTING.md Co-authored-by: Johannes Gäßler * improve * trailing space * Apply suggestions from code review Co-authored-by: Johannes Gäßler * update --------- Co-authored-by: Johannes Gäßler ") [ggml-org#18388](https://github.com/ggml-org/llama.cpp/pull/18388) [)](https://github.com/thecodacus/llama.cpp/commit/3595ae5963f1583f53beecf9725c919d309e15da "contributing: tighten AI usage policy (#18388) * contributing: tighten AI usage policy * refactor AGENTS.md * proofreading * update contributing * add claude.md * add trailing newline * add note about dishonest practices * rm point about dishonest * rm requirement watermarking * add .gemini/settings.json * allow initially AI-generated content * revise * Update CONTRIBUTING.md Co-authored-by: Johannes Gäßler * improve * trailing space * Apply suggestions from code review Co-authored-by: Johannes Gäßler * update --------- Co-authored-by: Johannes Gäßler ") | 7 months agoDec 29, 2025 |
| [.github](https://github.com/thecodacus/llama.cpp/tree/master/.github ".github") | [.github](https://github.com/thecodacus/llama.cpp/tree/master/.github ".github") | [gguf-py : bump version to 0.19.0 (](https://github.com/thecodacus/llama.cpp/commit/a290ce626663dae1d54f70bce3ca6d8f67aab62f "gguf-py : bump version to 0.19.0 (#22664) * gguf-py : bump version to 0.19.0 * bump poetry --------- Co-authored-by: Sigbjørn Skjæret ") [ggml-org#22664](https://github.com/ggml-org/llama.cpp/pull/22664) [)](https://github.com/thecodacus/llama.cpp/commit/a290ce626663dae1d54f70bce3ca6d8f67aab62f "gguf-py : bump version to 0.19.0 (#22664) * gguf-py : bump version to 0.19.0 * bump poetry --------- Co-authored-by: Sigbjørn Skjæret ") | 2 months agoMay 6, 2026 |
| [.pi/gg](https://github.com/thecodacus/llama.cpp/tree/master/.pi/gg "This path skips through empty directories") | [.pi/gg](https://github.com/thecodacus/llama.cpp/tree/master/.pi/gg "This path skips through empty directories") | [docs : update speculative decoding parameters after refactor (](https://github.com/thecodacus/llama.cpp/commit/846262d7875dcabf502a150fa3d7b9c770dde7eb "docs : update speculative decoding parameters after refactor (#22397) (#22539) * docs : update speculative decoding parameters after refactor (#22397) Update docs/speculative.md to reflect the new parameter naming scheme introduced in PR #22397: - Replace --draft-max/--draft-min with --spec-draft-n-max/--spec-draft-n-min - Replace --spec-ngram-size-n/m with per-implementation variants - Add documentation for all new --spec-ngram-*- parameters - Update all example commands Assisted-by: llama.cpp:local pi * pi : add rule to use gh CLI for GitHub resources Assisted-by: llama.cpp:local pi * docs : run llama-gen-docs * arg : fix typo") [ggml-or…](https://github.com/ggml-org/llama.cpp/pull/22397) | 2 months agoMay 4, 2026 |
| [benches](https://github.com/thecodacus/llama.cpp/tree/master/benches "benches") | [benches](https://github.com/thecodacus/llama.cpp/tree/master/benches "benches") | [benches : add Nemotron 3 Nano on DGX Spark (](https://github.com/thecodacus/llama.cpp/commit/9b342d0a9f2f4892daec065491583ec2be129685 "benches : add Nemotron 3 Nano on DGX Spark (#20652) [no ci]") [ggml-org#20652](https://github.com/ggml-org/llama.cpp/pull/20652) [)](https://github.com/thecodacus/llama.cpp/commit/9b342d0a9f2f4892daec065491583ec2be129685 "benches : add Nemotron 3 Nano on DGX Spark (#20652) [no ci]") | 4 months agoMar 16, 2026 |
| [ci](https://github.com/thecodacus/llama.cpp/tree/master/ci "ci") | [ci](https://github.com/thecodacus/llama.cpp/tree/master/ci "ci") | [ci: Add Windows Vulkan backend testing on Intel (](https://github.com/thecodacus/llama.cpp/commit/e43970099269b5b6da36b8977ad47697602e4e54 "ci: Add Windows Vulkan backend testing on Intel (#21292) * experimenting CI * Experimenting CI fix for MinGW * experimenting CI on Windows * modified script for integration with VisualStudio * added proxy handling * adding python version for Windows execution * fix iterator::end() dereference * fixed proxy handling * Fix errors occurring on Windows * fixed ci script * Reverted to master * Stripping test items to simplify Windows test * adjusting script for windows testing * Changed shell * Fixed shell * Fixed shell * Fix CI setting * Fix CI setting * Fix CI setting * Experimenting ci fix * Experimenting ci fix * Experimenting ci fix * Experimenting ci fix * experimenting fix for unit test error * Changed to use BUILD_LOW_PERF to skip python tests * Fix CI * Added option to specify Ninja generator * Reverted proxy related changes") [ggml-org#21292](https://github.com/ggml-org/llama.cpp/pull/21292) [)](https://github.com/thecodacus/llama.cpp/commit/e43970099269b5b6da36b8977ad47697602e4e54 "ci: Add Windows Vulkan backend testing on Intel (#21292) * experimenting CI * Experimenting CI fix for MinGW * experimenting CI on Windows * modified script for integration with VisualStudio * added proxy handling * adding python version for Windows execution * fix iterator::end() dereference * fixed proxy handling * Fix errors occurring on Windows * fixed ci script * Reverted to master * Stripping test items to simplify Windows test * adjusting script for windows testing * Changed shell * Fixed shell * Fixed shell * Fix CI setting * Fix CI setting * Fix CI setting * Experimenting ci fix * Experimenting ci fix * Experimenting ci fix * Experimenting ci fix * experimenting fix for unit test error * Changed to use BUILD_LOW_PERF to skip python tests * Fix CI * Added option to specify Ninja generator * Reverted proxy related changes") | 3 months agoApr 3, 2026 |
| [cmake](https://github.com/thecodacus/llama.cpp/tree/master/cmake "cmake") | [cmake](https://github.com/thecodacus/llama.cpp/tree/master/cmake "cmake") | [hexagon: add support for linux on snapdragon (](https://github.com/thecodacus/llama.cpp/commit/81069a808a29ae3048231e8b8c69758239f8e050 "hexagon: add support for linux on snapdragon (#21707) * hexagon: add support for debian on ex2 * hexagon: add -fvectotize to c/c++ cmake flags * hexagon: remove trailing white space * update onboarding steps * hexagon: update linux setup documentation * hexagon: update intallation scripts * Hexagon: update docs * hexagon: update onboarding scripts --------- Co-authored-by: Zack Li ") [ggml-org#21707](https://github.com/ggml-org/llama.cpp/pull/21707) [)](https://github.com/thecodacus/llama.cpp/commit/81069a808a29ae3048231e8b8c69758239f8e050 "hexagon: add support for linux on snapdragon (#21707) * hexagon: add support for debian on ex2 * hexagon: add -fvectotize to c/c++ cmake flags * hexagon: remove trailing white space * update onboarding steps * hexagon: update linux setup documentation * hexagon: update intallation scripts * Hexagon: update docs * hexagon: update onboarding scripts --------- Co-authored-by: Zack Li ") | 3 months agoApr 10, 2026 |
| [common](https://github.com/thecodacus/llama.cpp/tree/master/common "common") | [common](https://github.com/thecodacus/llama.cpp/tree/master/common "common") | [common: do not fit to unknown device memory (](https://github.com/thecodacus/llama.cpp/commit/a0101225bc60908572ed8b4dfd2ee1a61c69b689 "common: do not fit to unknown device memory (#22614) * common: do not fit to unknown device memory Signed-off-by: Florian Reinle * common: preserve host fallback for non-GPU fit devices Signed-off-by: Florian Reinle * common: keep unknown GPU fit memory at zero Signed-off-by: Florian Reinle --------- Signed-off-by: Florian Reinle ") [ggml-org#22614](https://github.com/ggml-org/llama.cpp/pull/22614) [)](https://github.com/thecodacus/llama.cpp/commit/a0101225bc60908572ed8b4dfd2ee1a61c69b689 "common: do not fit to unknown device memory (#22614) * common: do not fit to unknown device memory Signed-off-by: Florian Reinle * common: preserve host fallback for non-GPU fit devices Signed-off-by: Florian Reinle * common: keep unknown GPU fit memory at zero Signed-off-by: Florian Reinle --------- Signed-off-by: Florian Reinle ") | 2 months agoMay 6, 2026 |
| [docs](https://github.com/thecodacus/llama.cpp/tree/master/docs "docs") | [docs](https://github.com/thecodacus/llama.cpp/tree/master/docs "docs") | [mtmd : support MiniCPM-V 4.6 (](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") [ggml-org#22529](https://github.com/ggml-org/llama.cpp/pull/22529) [)](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") | 2 months agoMay 6, 2026 |
| [examples](https://github.com/thecodacus/llama.cpp/tree/master/examples "examples") | [examples](https://github.com/thecodacus/llama.cpp/tree/master/examples "examples") | [common : only load backends when required (](https://github.com/thecodacus/llama.cpp/commit/bf76ac77bed3f53a6d35968019ee34078bebf701 "common : only load backends when required (#22290) * common : only load backends when required Signed-off-by: Adrien Gallouët * llama : call ggml_backend_load_all() directly from llama_backend_init() Signed-off-by: Adrien Gallouët * Add ggml_backend_load_all() where llama_backend_init() is not used Signed-off-by: Adrien Gallouët --------- Signed-off-by: Adrien Gallouët ") [ggml-org#22290](https://github.com/ggml-org/llama.cpp/pull/22290) [)](https://github.com/thecodacus/llama.cpp/commit/bf76ac77bed3f53a6d35968019ee34078bebf701 "common : only load backends when required (#22290) * common : only load backends when required Signed-off-by: Adrien Gallouët * llama : call ggml_backend_load_all() directly from llama_backend_init() Signed-off-by: Adrien Gallouët * Add ggml_backend_load_all() where llama_backend_init() is not used Signed-off-by: Adrien Gallouët --------- Signed-off-by: Adrien Gallouët ") | 2 months agoMay 5, 2026 |
| [ggml](https://github.com/thecodacus/llama.cpp/tree/master/ggml "ggml") | [ggml](https://github.com/thecodacus/llama.cpp/tree/master/ggml "ggml") | [ggml-cpu: fuse RMS\_NORM + MUL on CPU backend (](https://github.com/thecodacus/llama.cpp/commit/f08f20a0e330da9530ded3e800dff9fbb05a3288 "ggml-cpu: fuse RMS_NORM + MUL on CPU backend (#22423)") [ggml-org#22423](https://github.com/ggml-org/llama.cpp/pull/22423) [)](https://github.com/thecodacus/llama.cpp/commit/f08f20a0e330da9530ded3e800dff9fbb05a3288 "ggml-cpu: fuse RMS_NORM + MUL on CPU backend (#22423)") | 2 months agoMay 6, 2026 |
| [gguf-py](https://github.com/thecodacus/llama.cpp/tree/master/gguf-py "gguf-py") | [gguf-py](https://github.com/thecodacus/llama.cpp/tree/master/gguf-py "gguf-py") | [mtmd : support MiniCPM-V 4.6 (](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") [ggml-org#22529](https://github.com/ggml-org/llama.cpp/pull/22529) [)](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") | 2 months agoMay 6, 2026 |
| [grammars](https://github.com/thecodacus/llama.cpp/tree/master/grammars "grammars") | [grammars](https://github.com/thecodacus/llama.cpp/tree/master/grammars "grammars") | [docs : document that JSON Schema is not available to model when using…](https://github.com/thecodacus/llama.cpp/commit/6e0c8cbc40c4abf49e5c52f0f51267c2afdfc053 "docs : document that JSON Schema is not available to model when using response_format (#18492) * Document unsupported JSON Schema annotations Add note about unsupported JSON Schema annotations. * Update README.md * Update README.md * Update README.md") | 7 months agoDec 30, 2025 |
| [include](https://github.com/thecodacus/llama.cpp/tree/master/include "include") | [include](https://github.com/thecodacus/llama.cpp/tree/master/include "include") | [llama : add option to save memory in device buffers (](https://github.com/thecodacus/llama.cpp/commit/d6e7b033a42b2880496d1e9db1df5527e48a8869 "llama : add option to save memory in device buffers (#22679) * llama : add option to save memory in device buffers * tests : extend llama-save-load-state") [ggml-org#22679](https://github.com/ggml-org/llama.cpp/pull/22679) [)](https://github.com/thecodacus/llama.cpp/commit/d6e7b033a42b2880496d1e9db1df5527e48a8869 "llama : add option to save memory in device buffers (#22679) * llama : add option to save memory in device buffers * tests : extend llama-save-load-state") | 2 months agoMay 4, 2026 |
| [licenses](https://github.com/thecodacus/llama.cpp/tree/master/licenses "licenses") | [licenses](https://github.com/thecodacus/llama.cpp/tree/master/licenses "licenses") | [refactor : remove libcurl, use OpenSSL when available (](https://github.com/thecodacus/llama.cpp/commit/516a4ca9b5f2fa72c2a71f412929a67cf76a6213 "refactor : remove libcurl, use OpenSSL when available (#18828)") [ggml-org#18828](https://github.com/ggml-org/llama.cpp/pull/18828) [)](https://github.com/thecodacus/llama.cpp/commit/516a4ca9b5f2fa72c2a71f412929a67cf76a6213 "refactor : remove libcurl, use OpenSSL when available (#18828)") | 6 months agoJan 14, 2026 |
| [media](https://github.com/thecodacus/llama.cpp/tree/master/media "media") | [media](https://github.com/thecodacus/llama.cpp/tree/master/media "media") | [media : add transparent icon svg and png \[no ci\] (](https://github.com/thecodacus/llama.cpp/commit/2cfef4d117d67ab1dec002915b48a15d11ee1973 "media : add transparent icon svg and png [no ci] (#15891)") [ggml-org#15891](https://github.com/ggml-org/llama.cpp/pull/15891) [)](https://github.com/thecodacus/llama.cpp/commit/2cfef4d117d67ab1dec002915b48a15d11ee1973 "media : add transparent icon svg and png [no ci] (#15891)") | 10 months agoSep 10, 2025 |
| [models](https://github.com/thecodacus/llama.cpp/tree/master/models "models") | [models](https://github.com/thecodacus/llama.cpp/tree/master/models "models") | [autoparser: support case of JSON\_NATIVE with per-call markers (test c…](https://github.com/thecodacus/llama.cpp/commit/e1a9a6dcbefccb4b864d9385ce8494f2a7f2ffcd "autoparser: support case of JSON_NATIVE with per-call markers (test case: Reka-Edge) (#21892)") | 3 months agoApr 15, 2026 |
| [pocs](https://github.com/thecodacus/llama.cpp/tree/master/pocs "pocs") | [pocs](https://github.com/thecodacus/llama.cpp/tree/master/pocs "pocs") | [libs : rename libcommon -> libllama-common (](https://github.com/thecodacus/llama.cpp/commit/6990e2f1f7581d332a6a1f34d6c567be70138184 "libs : rename libcommon -> libllama-common (#21936) * cmake : allow libcommon to be shared * cmake : rename libcommon to libllama-common * cont : set -fPIC for httplib * cont : export all symbols * cont : fix build_info exports * libs : add libllama-common-base * log : add common_log_get_verbosity_thold()") [ggml-org#21936](https://github.com/ggml-org/llama.cpp/pull/21936) [)](https://github.com/thecodacus/llama.cpp/commit/6990e2f1f7581d332a6a1f34d6c567be70138184 "libs : rename libcommon -> libllama-common (#21936) * cmake : allow libcommon to be shared * cmake : rename libcommon to libllama-common * cont : set -fPIC for httplib * cont : export all symbols * cont : fix build_info exports * libs : add libllama-common-base * log : add common_log_get_verbosity_thold()") | 3 months agoApr 17, 2026 |
| [requirements](https://github.com/thecodacus/llama.cpp/tree/master/requirements "requirements") | [requirements](https://github.com/thecodacus/llama.cpp/tree/master/requirements "requirements") | [py : Bump typer to latest to fix huggingface\_hub issue (](https://github.com/thecodacus/llama.cpp/commit/660386f6f88764afa615b9be0bfae4961fa115f1 "py : Bump typer to latest to fix huggingface_hub issue (#21701)") [ggml-org#21701](https://github.com/ggml-org/llama.cpp/pull/21701) [)](https://github.com/thecodacus/llama.cpp/commit/660386f6f88764afa615b9be0bfae4961fa115f1 "py : Bump typer to latest to fix huggingface_hub issue (#21701)") | 3 months agoApr 11, 2026 |
| [scripts](https://github.com/thecodacus/llama.cpp/tree/master/scripts "scripts") | [scripts](https://github.com/thecodacus/llama.cpp/tree/master/scripts "scripts") | [sync : ggml](https://github.com/thecodacus/llama.cpp/commit/70a8309114c4f6df8fb53517f6c61268160a76c8 "sync : ggml") | 2 months agoMay 5, 2026 |
| [src](https://github.com/thecodacus/llama.cpp/tree/master/src "src") | [src](https://github.com/thecodacus/llama.cpp/tree/master/src "src") | [model : don't crash on unsupported architecture (](https://github.com/thecodacus/llama.cpp/commit/5207d120eac2393fdad6328b44dbcbfc5dea20e4 "model : don't crash on unsupported architecture (#22742) * model: don't crash on unsupported architecture * Update src/llama-model.cpp Co-authored-by: Sigbjørn Skjæret --------- Co-authored-by: Sigbjørn Skjæret ") [ggml-org#22742](https://github.com/ggml-org/llama.cpp/pull/22742) [)](https://github.com/thecodacus/llama.cpp/commit/5207d120eac2393fdad6328b44dbcbfc5dea20e4 "model : don't crash on unsupported architecture (#22742) * model: don't crash on unsupported architecture * Update src/llama-model.cpp Co-authored-by: Sigbjørn Skjæret --------- Co-authored-by: Sigbjørn Skjæret ") | 2 months agoMay 6, 2026 |
| [tests](https://github.com/thecodacus/llama.cpp/tree/master/tests "tests") | [tests](https://github.com/thecodacus/llama.cpp/tree/master/tests "tests") | [common : only load backends when required (](https://github.com/thecodacus/llama.cpp/commit/bf76ac77bed3f53a6d35968019ee34078bebf701 "common : only load backends when required (#22290) * common : only load backends when required Signed-off-by: Adrien Gallouët * llama : call ggml_backend_load_all() directly from llama_backend_init() Signed-off-by: Adrien Gallouët * Add ggml_backend_load_all() where llama_backend_init() is not used Signed-off-by: Adrien Gallouët --------- Signed-off-by: Adrien Gallouët ") [ggml-org#22290](https://github.com/ggml-org/llama.cpp/pull/22290) [)](https://github.com/thecodacus/llama.cpp/commit/bf76ac77bed3f53a6d35968019ee34078bebf701 "common : only load backends when required (#22290) * common : only load backends when required Signed-off-by: Adrien Gallouët * llama : call ggml_backend_load_all() directly from llama_backend_init() Signed-off-by: Adrien Gallouët * Add ggml_backend_load_all() where llama_backend_init() is not used Signed-off-by: Adrien Gallouët --------- Signed-off-by: Adrien Gallouët ") | 2 months agoMay 5, 2026 |
| [tools](https://github.com/thecodacus/llama.cpp/tree/master/tools "tools") | [tools](https://github.com/thecodacus/llama.cpp/tree/master/tools "tools") | [mtmd : support MiniCPM-V 4.6 (](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") [ggml-org#22529](https://github.com/ggml-org/llama.cpp/pull/22529) [)](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") | 2 months agoMay 6, 2026 |
| [vendor](https://github.com/thecodacus/llama.cpp/tree/master/vendor "vendor") | [vendor](https://github.com/thecodacus/llama.cpp/tree/master/vendor "vendor") | [vendor : update cpp-httplib to 0.43.3 (](https://github.com/thecodacus/llama.cpp/commit/a09a00e50259578ec0127d5116cb744191efeaac "vendor : update cpp-httplib to 0.43.3 (#22686)") [ggml-org#22686](https://github.com/ggml-org/llama.cpp/pull/22686) [)](https://github.com/thecodacus/llama.cpp/commit/a09a00e50259578ec0127d5116cb744191efeaac "vendor : update cpp-httplib to 0.43.3 (#22686)") | 2 months agoMay 5, 2026 |
| [.clang-format](https://github.com/thecodacus/llama.cpp/blob/master/.clang-format ".clang-format") | [.clang-format](https://github.com/thecodacus/llama.cpp/blob/master/.clang-format ".clang-format") | [fix: apply clang-format to CUDA macros (](https://github.com/thecodacus/llama.cpp/commit/f1fbffb5c0b34b2a68febb7da3fd0f8333f1ed4c "fix: apply clang-format to CUDA macros (#16017) clang-format previously broke long CUDA macros (e.g. __launch_bounds__) into unreadable line breaks inside template declarations, such as: template __launch_bounds__(nwarps*ggml_cuda_get_physical_warp_size(), 1) This change adjusts formatting rules so that CUDA macros remain consistent and aligned with the surrounding template syntax.") [ggml-org#16017](https://github.com/ggml-org/llama.cpp/pull/16017) [)](https://github.com/thecodacus/llama.cpp/commit/f1fbffb5c0b34b2a68febb7da3fd0f8333f1ed4c "fix: apply clang-format to CUDA macros (#16017) clang-format previously broke long CUDA macros (e.g. __launch_bounds__) into unreadable line breaks inside template declarations, such as: template __launch_bounds__(nwarps*ggml_cuda_get_physical_warp_size(), 1) This change adjusts formatting rules so that CUDA macros remain consistent and aligned with the surrounding template syntax.") | 10 months agoSep 16, 2025 |
| [.clang-tidy](https://github.com/thecodacus/llama.cpp/blob/master/.clang-tidy ".clang-tidy") | [.clang-tidy](https://github.com/thecodacus/llama.cpp/blob/master/.clang-tidy ".clang-tidy") | [clang-tidy : disable warning about performance enum size (](https://github.com/thecodacus/llama.cpp/commit/351f3da39c85f59d581fc184f09283da7f099a3b "clang-tidy : disable warning about performance enum size (#16127) Disable 'performance-enum-size' checking: Enum 'llama_token_type' uses a larger base type ('unsigned int', size: 4 bytes) than necessary for its value set, consider using 'std::uint8_t' (1 byte) as the base type to reduce its size.") [ggml-org#16127](https://github.com/ggml-org/llama.cpp/pull/16127) | 10 months agoSep 22, 2025 |
| [.dockerignore](https://github.com/thecodacus/llama.cpp/blob/master/.dockerignore ".dockerignore") | [.dockerignore](https://github.com/thecodacus/llama.cpp/blob/master/.dockerignore ".dockerignore") | [ci : fix docker build number and tag name (](https://github.com/thecodacus/llama.cpp/commit/ea9c32be71b91b42ecc538bd902e93cbb5fb36cb "ci : fix docker build number and tag name (#9638) * ci : fix docker build number and tag name * fine-grant permissions") [ggml-org#9638](https://github.com/ggml-org/llama.cpp/pull/9638) [)](https://github.com/thecodacus/llama.cpp/commit/ea9c32be71b91b42ecc538bd902e93cbb5fb36cb "ci : fix docker build number and tag name (#9638) * ci : fix docker build number and tag name * fine-grant permissions") | 2 years agoSep 25, 2024 |
| [.ecrc](https://github.com/thecodacus/llama.cpp/blob/master/.ecrc ".ecrc") | [.ecrc](https://github.com/thecodacus/llama.cpp/blob/master/.ecrc ".ecrc") | [common : Update stb\_image.h to latest version (](https://github.com/thecodacus/llama.cpp/commit/ad76569f8e78ab6ca921bda25cef25a157361719 "common : Update stb_image.h to latest version (#9161) * Update stb_image.h to latest version Fixes https://github.com/ggerganov/llama.cpp/issues/7431 * Update .ecrc") [ggml-org#9161](https://github.com/ggml-org/llama.cpp/pull/9161) [)](https://github.com/thecodacus/llama.cpp/commit/ad76569f8e78ab6ca921bda25cef25a157361719 "common : Update stb_image.h to latest version (#9161) * Update stb_image.h to latest version Fixes https://github.com/ggerganov/llama.cpp/issues/7431 * Update .ecrc") | 2 years agoAug 27, 2024 |
| [.editorconfig](https://github.com/thecodacus/llama.cpp/blob/master/.editorconfig ".editorconfig") | [.editorconfig](https://github.com/thecodacus/llama.cpp/blob/master/.editorconfig ".editorconfig") | [server: (webui) no more gzip compression (](https://github.com/thecodacus/llama.cpp/commit/4a00bbfed62a9714bc94a3431e6cc37615b175a2 "server: (webui) no more gzip compression (#21073) * webui: no more gzip * try changing a small line * Revert \"try changing a small line\" This reverts commit 0d7a3531593d87b724d404c8727a96becab3ab07. * fix lint * fix test * rebuild * split into html/css/js * lint * chore: update webui build output * chore: Update git hooks script * server: update webui build output * chore: Update pre-commit hook * refactor: Cleanup --------- Co-authored-by: Aleksander Grygier ") [ggml-org#21073](https://github.com/ggml-org/llama.cpp/pull/21073) [)](https://github.com/thecodacus/llama.cpp/commit/4a00bbfed62a9714bc94a3431e6cc37615b175a2 "server: (webui) no more gzip compression (#21073) * webui: no more gzip * try changing a small line * Revert \"try changing a small line\" This reverts commit 0d7a3531593d87b724d404c8727a96becab3ab07. * fix lint * fix test * rebuild * split into html/css/js * lint * chore: update webui build output * chore: Update git hooks script * server: update webui build output * chore: Update pre-commit hook * refactor: Cleanup --------- Co-authored-by: Aleksander Grygier ") | 4 months agoMar 31, 2026 |
| [.flake8](https://github.com/thecodacus/llama.cpp/blob/master/.flake8 ".flake8") | [.flake8](https://github.com/thecodacus/llama.cpp/blob/master/.flake8 ".flake8") | [llama : move end-user examples to tools directory (](https://github.com/thecodacus/llama.cpp/commit/1d36b3670b285e69e58b9d687c770a2a0a192194 "llama : move end-user examples to tools directory (#13249) * llama : move end-user examples to tools directory --------- Co-authored-by: Xuan Son Nguyen ") [ggml-org#13249](https://github.com/ggml-org/llama.cpp/pull/13249) [)](https://github.com/thecodacus/llama.cpp/commit/1d36b3670b285e69e58b9d687c770a2a0a192194 "llama : move end-user examples to tools directory (#13249) * llama : move end-user examples to tools directory --------- Co-authored-by: Xuan Son Nguyen ") | last yearMay 2, 2025 |
| [.gitattributes](https://github.com/thecodacus/llama.cpp/blob/master/.gitattributes ".gitattributes") | [.gitattributes](https://github.com/thecodacus/llama.cpp/blob/master/.gitattributes ".gitattributes") | [server: (webui) no more gzip compression (](https://github.com/thecodacus/llama.cpp/commit/4a00bbfed62a9714bc94a3431e6cc37615b175a2 "server: (webui) no more gzip compression (#21073) * webui: no more gzip * try changing a small line * Revert \"try changing a small line\" This reverts commit 0d7a3531593d87b724d404c8727a96becab3ab07. * fix lint * fix test * rebuild * split into html/css/js * lint * chore: update webui build output * chore: Update git hooks script * server: update webui build output * chore: Update pre-commit hook * refactor: Cleanup --------- Co-authored-by: Aleksander Grygier ") [ggml-org#21073](https://github.com/ggml-org/llama.cpp/pull/21073) [)](https://github.com/thecodacus/llama.cpp/commit/4a00bbfed62a9714bc94a3431e6cc37615b175a2 "server: (webui) no more gzip compression (#21073) * webui: no more gzip * try changing a small line * Revert \"try changing a small line\" This reverts commit 0d7a3531593d87b724d404c8727a96becab3ab07. * fix lint * fix test * rebuild * split into html/css/js * lint * chore: update webui build output * chore: Update git hooks script * server: update webui build output * chore: Update pre-commit hook * refactor: Cleanup --------- Co-authored-by: Aleksander Grygier ") | 4 months agoMar 31, 2026 |
| [.gitignore](https://github.com/thecodacus/llama.cpp/blob/master/.gitignore ".gitignore") | [.gitignore](https://github.com/thecodacus/llama.cpp/blob/master/.gitignore ".gitignore") | [feat: migrate to PEP 621 and add uv support (](https://github.com/thecodacus/llama.cpp/commit/750141969ca98d91be460538655904591f337b82 "feat: migrate to PEP 621 and add uv support (#21907) * feat: migrate to PEP 621 and add uv support * fix: remove upper bound on protobuf * remove poetry.lock and uv.lock * fix/add torch dependency version and markers * fix dev-dependency deprecation warning * gguf-py : update python version requirement to 3.10 --------- Co-authored-by: David Huggins-Daines Co-authored-by: Daniel Bevenius ") [ggml-org#21907](https://github.com/ggml-org/llama.cpp/pull/21907) [)](https://github.com/thecodacus/llama.cpp/commit/750141969ca98d91be460538655904591f337b82 "feat: migrate to PEP 621 and add uv support (#21907) * feat: migrate to PEP 621 and add uv support * fix: remove upper bound on protobuf * remove poetry.lock and uv.lock * fix/add torch dependency version and markers * fix dev-dependency deprecation warning * gguf-py : update python version requirement to 3.10 --------- Co-authored-by: David Huggins-Daines Co-authored-by: Daniel Bevenius ") | 2 months agoMay 6, 2026 |
| [.gitmodules](https://github.com/thecodacus/llama.cpp/blob/master/.gitmodules ".gitmodules") | [.gitmodules](https://github.com/thecodacus/llama.cpp/blob/master/.gitmodules ".gitmodules") | [ggml : remove kompute backend (](https://github.com/thecodacus/llama.cpp/commit/d4cdd9c1c3cebdafca735958597de4ff7b7c0f54 "ggml : remove kompute backend (#14501) ggml-ci") [ggml-org#14501](https://github.com/ggml-org/llama.cpp/pull/14501) [)](https://github.com/thecodacus/llama.cpp/commit/d4cdd9c1c3cebdafca735958597de4ff7b7c0f54 "ggml : remove kompute backend (#14501) ggml-ci") | last yearJul 3, 2025 |
| [.pre-commit-config.yaml](https://github.com/thecodacus/llama.cpp/blob/master/.pre-commit-config.yaml ".pre-commit-config.yaml") | [.pre-commit-config.yaml](https://github.com/thecodacus/llama.cpp/blob/master/.pre-commit-config.yaml ".pre-commit-config.yaml") | [convert.py : add python logging instead of print() (](https://github.com/thecodacus/llama.cpp/commit/a2ac89d6efb41b535778bfeaecaae8fe295b6ed3 "convert.py : add python logging instead of print() (#6511) * convert.py: add python logging instead of print() * convert.py: verbose flag takes priority over dump flag log suppression * convert.py: named instance logging * convert.py: use explicit logger id string * convert.py: convert extra print() to named logger * convert.py: sys.stderr.write --> logger.error * *.py: Convert all python scripts to use logging module * requirements.txt: remove extra line * flake8: update flake8 ignore and exclude to match ci settings * gh-actions: add flake8-no-print to flake8 lint step * pre-commit: add flake8-no-print to flake8 and also update pre-commit version * convert-hf-to-gguf.py: print() to logger conversion * *.py: logging basiconfig refactor to use conditional expression * *.py: removed commented out logging * fixup! *.py: logging basiconfig refactor to use conditional expression * constant.py: logger.error then exit should be a raise exception instead * *.py: Convert logger error and sys.exit() into a raise exception (for atypical error) * gguf-convert-endian.py: refactor convert_byteorder() to use tqdm progressbar * verify-checksum-model.py: This is the result of the program, it should be printed to stdout. * compare-llama-bench.py: add blank line for readability during missing repo response * reader.py: read_gguf_file() use print() over logging * convert.py: warning goes to stderr and won't hurt the dump output * gguf-dump.py: dump_metadata() should print to stdout * convert-hf-to-gguf.py: print --> logger.debug or ValueError() * verify-checksum-models.py: use print() for printing table * *.py: refactor logging.basicConfig() * gguf-py/gguf/*.py: use __name__ as logger name Since they will be imported and not run directly. * python-lint.yml: use .flake8 file instead * constants.py: logger no longer required * convert-hf-to-gguf.py: add additional logging * convert-hf-to-gguf.py: print() --> logger * *.py: fix flake8 warnings * revert changes to convert-hf-to-gguf.py for get_name() * convert-hf-to-gguf-update.py: use triple quoted f-string instead * *.py: accidentally corrected the wrong line * *.py: add compilade warning suggestions and style fixes") [ggml-org#6511](https://github.com/ggml-org/llama.cpp/pull/6511) [)](https://github.com/thecodacus/llama.cpp/commit/a2ac89d6efb41b535778bfeaecaae8fe295b6ed3 "convert.py : add python logging instead of print() (#6511) * convert.py: add python logging instead of print() * convert.py: verbose flag takes priority over dump flag log suppression * convert.py: named instance logging * convert.py: use explicit logger id string * convert.py: convert extra print() to named logger * convert.py: sys.stderr.write --> logger.error * *.py: Convert all python scripts to use logging module * requirements.txt: remove extra line * flake8: update flake8 ignore and exclude to match ci settings * gh-actions: add flake8-no-print to flake8 lint step * pre-commit: add flake8-no-print to flake8 and also update pre-commit version * convert-hf-to-gguf.py: print() to logger conversion * *.py: logging basiconfig refactor to use conditional expression * *.py: removed commented out logging * fixup! *.py: logging basiconfig refactor to use conditional expression * constant.py: logger.error then exit should be a raise exception instead * *.py: Convert logger error and sys.exit() into a raise exception (for atypical error) * gguf-convert-endian.py: refactor convert_byteorder() to use tqdm progressbar * verify-checksum-model.py: This is the result of the program, it should be printed to stdout. * compare-llama-bench.py: add blank line for readability during missing repo response * reader.py: read_gguf_file() use print() over logging * convert.py: warning goes to stderr and won't hurt the dump output * gguf-dump.py: dump_metadata() should print to stdout * convert-hf-to-gguf.py: print --> logger.debug or ValueError() * verify-checksum-models.py: use print() for printing table * *.py: refactor logging.basicConfig() * gguf-py/gguf/*.py: use __name__ as logger name Since they will be imported and not run directly. * python-lint.yml: use .flake8 file instead * constants.py: logger no longer required * convert-hf-to-gguf.py: add additional logging * convert-hf-to-gguf.py: print() --> logger * *.py: fix flake8 warnings * revert changes to convert-hf-to-gguf.py for get_name() * convert-hf-to-gguf-update.py: use triple quoted f-string instead * *.py: accidentally corrected the wrong line * *.py: add compilade warning suggestions and style fixes") | 2 years agoMay 3, 2024 |
| [AGENTS.md](https://github.com/thecodacus/llama.cpp/blob/master/AGENTS.md "AGENTS.md") | [AGENTS.md](https://github.com/thecodacus/llama.cpp/blob/master/AGENTS.md "AGENTS.md") | [contrib : rewrite AGENTS.md, make it more clear about project values (](https://github.com/thecodacus/llama.cpp/commit/c30e012253dd9e322c8e3424f808a5c74ecc46bf "contrib : rewrite AGENTS.md, make it more clear about project values (#21270) * contrib : rewrite AGENTS.md, make it more clear about types of permitted AI usage * permit AI for writing code") [g…](https://github.com/ggml-org/llama.cpp/pull/21270) | 4 months agoApr 1, 2026 |
| [AUTHORS](https://github.com/thecodacus/llama.cpp/blob/master/AUTHORS "AUTHORS") | [AUTHORS](https://github.com/thecodacus/llama.cpp/blob/master/AUTHORS "AUTHORS") | [authors : update (](https://github.com/thecodacus/llama.cpp/commit/1239267cc4e5a1c9fc6546825eefe13c856e7458 "authors : update (#19263) [no ci]") [ggml-org#19263](https://github.com/ggml-org/llama.cpp/pull/19263) [)](https://github.com/thecodacus/llama.cpp/commit/1239267cc4e5a1c9fc6546825eefe13c856e7458 "authors : update (#19263) [no ci]") | 5 months agoFeb 2, 2026 |
| [CLAUDE.md](https://github.com/thecodacus/llama.cpp/blob/master/CLAUDE.md "CLAUDE.md") | [CLAUDE.md](https://github.com/thecodacus/llama.cpp/blob/master/CLAUDE.md "CLAUDE.md") | [contributing: tighten AI usage policy (](https://github.com/thecodacus/llama.cpp/commit/3595ae5963f1583f53beecf9725c919d309e15da "contributing: tighten AI usage policy (#18388) * contributing: tighten AI usage policy * refactor AGENTS.md * proofreading * update contributing * add claude.md * add trailing newline * add note about dishonest practices * rm point about dishonest * rm requirement watermarking * add .gemini/settings.json * allow initially AI-generated content * revise * Update CONTRIBUTING.md Co-authored-by: Johannes Gäßler * improve * trailing space * Apply suggestions from code review Co-authored-by: Johannes Gäßler * update --------- Co-authored-by: Johannes Gäßler ") [ggml-org#18388](https://github.com/ggml-org/llama.cpp/pull/18388) [)](https://github.com/thecodacus/llama.cpp/commit/3595ae5963f1583f53beecf9725c919d309e15da "contributing: tighten AI usage policy (#18388) * contributing: tighten AI usage policy * refactor AGENTS.md * proofreading * update contributing * add claude.md * add trailing newline * add note about dishonest practices * rm point about dishonest * rm requirement watermarking * add .gemini/settings.json * allow initially AI-generated content * revise * Update CONTRIBUTING.md Co-authored-by: Johannes Gäßler * improve * trailing space * Apply suggestions from code review Co-authored-by: Johannes Gäßler * update --------- Co-authored-by: Johannes Gäßler ") | 7 months agoDec 29, 2025 |
| [CMakeLists.txt](https://github.com/thecodacus/llama.cpp/blob/master/CMakeLists.txt "CMakeLists.txt") | [CMakeLists.txt](https://github.com/thecodacus/llama.cpp/blob/master/CMakeLists.txt "CMakeLists.txt") | [libs : rename libcommon -> libllama-common (](https://github.com/thecodacus/llama.cpp/commit/6990e2f1f7581d332a6a1f34d6c567be70138184 "libs : rename libcommon -> libllama-common (#21936) * cmake : allow libcommon to be shared * cmake : rename libcommon to libllama-common * cont : set -fPIC for httplib * cont : export all symbols * cont : fix build_info exports * libs : add libllama-common-base * log : add common_log_get_verbosity_thold()") [ggml-org#21936](https://github.com/ggml-org/llama.cpp/pull/21936) [)](https://github.com/thecodacus/llama.cpp/commit/6990e2f1f7581d332a6a1f34d6c567be70138184 "libs : rename libcommon -> libllama-common (#21936) * cmake : allow libcommon to be shared * cmake : rename libcommon to libllama-common * cont : set -fPIC for httplib * cont : export all symbols * cont : fix build_info exports * libs : add libllama-common-base * log : add common_log_get_verbosity_thold()") | 3 months agoApr 17, 2026 |
| [CMakePresets.json](https://github.com/thecodacus/llama.cpp/blob/master/CMakePresets.json "CMakePresets.json") | [CMakePresets.json](https://github.com/thecodacus/llama.cpp/blob/master/CMakePresets.json "CMakePresets.json") | [cmake : Add CMake presets for Linux and GCC (](https://github.com/thecodacus/llama.cpp/commit/84b396e0510855a95d591afdf1f21c562cb3712a "cmake : Add CMake presets for Linux and GCC (#14656)") [ggml-org#14656](https://github.com/ggml-org/llama.cpp/pull/14656) [)](https://github.com/thecodacus/llama.cpp/commit/84b396e0510855a95d591afdf1f21c562cb3712a "cmake : Add CMake presets for Linux and GCC (#14656)") | last yearJul 13, 2025 |
| [CODEOWNERS](https://github.com/thecodacus/llama.cpp/blob/master/CODEOWNERS "CODEOWNERS") | [CODEOWNERS](https://github.com/thecodacus/llama.cpp/blob/master/CODEOWNERS "CODEOWNERS") | [Github: set meta backend code owner (](https://github.com/thecodacus/llama.cpp/commit/7ec36aa861d0acfed7a57add0b5e17b92aa40ffd "Github: set meta backend code owner (#22388)") [ggml-org#22388](https://github.com/ggml-org/llama.cpp/pull/22388) [)](https://github.com/thecodacus/llama.cpp/commit/7ec36aa861d0acfed7a57add0b5e17b92aa40ffd "Github: set meta backend code owner (#22388)") | 3 months agoApr 26, 2026 |
| [CONTRIBUTING.md](https://github.com/thecodacus/llama.cpp/blob/master/CONTRIBUTING.md "CONTRIBUTING.md") | [CONTRIBUTING.md](https://github.com/thecodacus/llama.cpp/blob/master/CONTRIBUTING.md "CONTRIBUTING.md") | [docs : explicit about banning accounts that violates policy (](https://github.com/thecodacus/llama.cpp/commit/568aec82d2fc48341c54cae565768ac75072a31d "docs : explicit about banning accounts that violates policy (#19593)") [ggml-org…](https://github.com/ggml-org/llama.cpp/pull/19593) | 4 months agoMar 21, 2026 |
| [LICENSE](https://github.com/thecodacus/llama.cpp/blob/master/LICENSE "LICENSE") | [LICENSE](https://github.com/thecodacus/llama.cpp/blob/master/LICENSE "LICENSE") | [docs : Minor cleanups (](https://github.com/thecodacus/llama.cpp/commit/7a4ca3cbd905907ac4d12bc14b878fdfbe4fd1d6 "docs : Minor cleanups (#19252) * Update old URLs to github.com/ggml-org/ * Bump copyrights") [ggml-org#19252](https://github.com/ggml-org/llama.cpp/pull/19252) [)](https://github.com/thecodacus/llama.cpp/commit/7a4ca3cbd905907ac4d12bc14b878fdfbe4fd1d6 "docs : Minor cleanups (#19252) * Update old URLs to github.com/ggml-org/ * Bump copyrights") | 5 months agoFeb 2, 2026 |
| [Makefile](https://github.com/thecodacus/llama.cpp/blob/master/Makefile "Makefile") | [Makefile](https://github.com/thecodacus/llama.cpp/blob/master/Makefile "Makefile") | [make : remove make in favor of CMake (](https://github.com/thecodacus/llama.cpp/commit/37f10f955f70e0158d50343d0b9a3f92d194daae "make : remove make in favor of CMake (#15449) This commit removes the content from the Makefile and updates the current deprecation message to information that `make` has been replaced by CMake instead. The message when `make` is invoked will now be the following: ```console $ make Makefile:6: *** Build system changed: The Makefile build has been replaced by CMake. For build instructions see: https://github.com/ggml-org/llama.cpp/blob/master/docs/build.md . Stop. ``` The motivation for this is that many, if not all targets fail to build now, after changes to the system, and `make` has also been deprected for some time now.") [ggml-org#15449](https://github.com/ggml-org/llama.cpp/pull/15449) [)](https://github.com/thecodacus/llama.cpp/commit/37f10f955f70e0158d50343d0b9a3f92d194daae "make : remove make in favor of CMake (#15449) This commit removes the content from the Makefile and updates the current deprecation message to information that `make` has been replaced by CMake instead. The message when `make` is invoked will now be the following: ```console $ make Makefile:6: *** Build system changed: The Makefile build has been replaced by CMake. For build instructions see: https://github.com/ggml-org/llama.cpp/blob/master/docs/build.md . Stop. ``` The motivation for this is that many, if not all targets fail to build now, after changes to the system, and `make` has also been deprected for some time now.") | 11 months agoAug 20, 2025 |
| [README.md](https://github.com/thecodacus/llama.cpp/blob/master/README.md "README.md") | [README.md](https://github.com/thecodacus/llama.cpp/blob/master/README.md "README.md") | [readme : clarify MODEL\_ENDPOINT usage (](https://github.com/thecodacus/llama.cpp/commit/c9dc43333f7b56fb7b1f306771f4a477602f6d81 "readme : clarify MODEL_ENDPOINT usage (#20941) Signed-off-by: Adrien Gallouët ") [ggml-org#20941](https://github.com/ggml-org/llama.cpp/pull/20941) [)](https://github.com/thecodacus/llama.cpp/commit/c9dc43333f7b56fb7b1f306771f4a477602f6d81 "readme : clarify MODEL_ENDPOINT usage (#20941) Signed-off-by: Adrien Gallouët ") | 4 months agoMar 24, 2026 |
| [SECURITY.md](https://github.com/thecodacus/llama.cpp/blob/master/SECURITY.md "SECURITY.md") | [SECURITY.md](https://github.com/thecodacus/llama.cpp/blob/master/SECURITY.md "SECURITY.md") | [docs : fix broken link and typo (](https://github.com/thecodacus/llama.cpp/commit/423cf0b26fc0b72ff4bb4656a68a607d38b95fe5 "docs : fix broken link and typo (#19560)") [ggml-org#19560](https://github.com/ggml-org/llama.cpp/pull/19560) [)](https://github.com/thecodacus/llama.cpp/commit/423cf0b26fc0b72ff4bb4656a68a607d38b95fe5 "docs : fix broken link and typo (#19560)") | 5 months agoFeb 13, 2026 |
| [build-xcframework.sh](https://github.com/thecodacus/llama.cpp/blob/master/build-xcframework.sh "build-xcframework.sh") | [build-xcframework.sh](https://github.com/thecodacus/llama.cpp/blob/master/build-xcframework.sh "build-xcframework.sh") | [build : remove LLAMA\_HTTPLIB option (](https://github.com/thecodacus/llama.cpp/commit/9e118b97c456d378dea49a8c46b850c71fc18363 "build : remove LLAMA_HTTPLIB option (#19623) This option was introduced as a workaround because cpp-httplib could not build on visionOS. Since it has been fixed and now compiles on all platforms, we can remove it and simplify many things. Signed-off-by: Adrien Gallouët ") [ggml-org#19623](https://github.com/ggml-org/llama.cpp/pull/19623) [)](https://github.com/thecodacus/llama.cpp/commit/9e118b97c456d378dea49a8c46b850c71fc18363 "build : remove LLAMA_HTTPLIB option (#19623) This option was introduced as a workaround because cpp-httplib could not build on visionOS. Since it has been fixed and now compiles on all platforms, we can remove it and simplify many things. Signed-off-by: Adrien Gallouët ") | 5 months agoFeb 15, 2026 |
| [convert\_hf\_to\_gguf.py](https://github.com/thecodacus/llama.cpp/blob/master/convert_hf_to_gguf.py "convert_hf_to_gguf.py") | [convert\_hf\_to\_gguf.py](https://github.com/thecodacus/llama.cpp/blob/master/convert_hf_to_gguf.py "convert_hf_to_gguf.py") | [mtmd : support MiniCPM-V 4.6 (](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") [ggml-org#22529](https://github.com/ggml-org/llama.cpp/pull/22529) [)](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") | 2 months agoMay 6, 2026 |
| [convert\_hf\_to\_gguf\_update.py](https://github.com/thecodacus/llama.cpp/blob/master/convert_hf_to_gguf_update.py "convert_hf_to_gguf_update.py") | [convert\_hf\_to\_gguf\_update.py](https://github.com/thecodacus/llama.cpp/blob/master/convert_hf_to_gguf_update.py "convert_hf_to_gguf_update.py") | [mtmd : support MiniCPM-V 4.6 (](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb Co-authored-by: Sigbjørn Skjæret ") [ggml-org#22529](https://github.com/ggml-org/llama.cpp/pull/22529) [)](https://github.com/thecodacus/llama.cpp/commit/2496f9c14965c39589f53eea31bdb6d762b1d360 "mtmd : support MiniCPM-V 4.6 (#22529) * Support MiniCPM-V 4.6 in new branch Signed-off-by: tc-mb * fix code bug Signed-off-by: tc-mb * fix pre-commit Signed-off-by: tc-mb * fix convert Signed-off-by: tc-mb * rename clip_graph_minicpmv4_6 Signed-off-by: tc-mb * use new TYPE_MINICPMV4_6 Signed-off-by: tc-mb * use build_attn to allow flash attention support Signed-off-by: tc-mb * no use legacy code, restored here. Signed-off-by: tc-mb * use the existing tensors name Signed-off-by: tc-mb * unused ctx->model.hparams.minicpmv_version Signed-off-by: tc-mb * use n_merge for slice alignment Signed-off-by: tc-mb * borrow wa_layer_indexes for vit_merger insertion point Signed-off-by: tc-mb * fix code style Signed-off-by: tc-mb * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret * use filter_tensors and add model.vision_tower Signed-off-by: tc-mb * fix chkhsh Signed-off-by: tc-mb * fix type check Signed-off-by: tc-mb --------- Signed-off-by: tc-mb