GitHub / abetlen/llama-cpp-python / commits
Python bindings for llama.cpp
| SHA | Message | Author | Date | Stats |
|---|---|---|---|---|
| fdf38b3e |
fix: avoid cleanup errors for partially initialized LlamaModel (#2173)
Co-authored-by: abetlen <a****n@g****m> |
usernames122 <8****2@u****m>
Committed by: GitHub <n****y@g****m> |
3 months ago | |
| 6bdab5d8 |
fix: suppress stdout and stderr in Jupyter notebooks (#2181)
Co-authored-by: Anai-Guo <2****o@u****m> |
Tai An <a****1@o****m>
Committed by: GitHub <n****y@g****m> |
3 months ago | |
| b91460b6 |
feat: enable arm64 musl builds (#2221)
Co-authored-by: abetlen <a****n@g****m> |
Alex O'Connell <3****6@u****m>
Committed by: GitHub <n****y@g****m> |
3 months ago | |
| f160bf7a |
Fix: model fails to load when chat template uses HuggingFace generation tags ...
Co-authored-by: abetlen <a****n@g****m> |
Tobias <5****2@u****m>
Committed by: GitHub <n****y@g****m> |
3 months ago | |
| 2c455a5b | feat: Update llama.cpp to d749821db (#2233) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
3 months ago | |
| 3bda0914 | docs: add contributing guide (#2229) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
3 months ago | |
| 52fe54bb | feat: Update llama.cpp to c0c7e147e (#2228) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
3 months ago | |
| 5dd9b1ce | feat: Update llama.cpp to b9a2170fc (#2223) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
3 months ago | |
| c7bea711 |
chore: migrate llama.cpp submodule to ggml-org (#2034)
Co-authored-by: abetlen <a****n@g****m> |
shalinib-ibm <S****i@i****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 7664a3ed | feat: Update llama.cpp to ggerganov/llama.cpp@91e84fed6 (#2218) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 95ccb191 |
fix(embedding): set kv_unified=True when embedding=True to enable batch proce...
Co-authored-by: abetlen <a****n@g****m> |
Sanjana Brahmbhatt <9****3@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 4a1a8ecd | chore: bump version to 0.3.23 (#2215) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 56841123 | feat: update llama.cpp to 7d442abf (#2214) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| f8c1f36b | fix(embed): mark all tokens as output to suppress llama.cpp 'overriding' INFO... |
Tai An <a****1@o****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| f7746900 | feat: update llama.cpp to 5d6f18a63 (#2207) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 128c331b | fix: configure n_seq_max for batched embeddings (#2206) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 90e8df95 | fix(_internals): use n_tokens0 offset when enabling last-token logits in add_... |
Tai An <a****1@o****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 14d7846f | fix(ci): skip unsupported Windows CUDA versions (#2204) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| bc6ff9f2 | fix(ci): install CUDA CCCL headers for wheel builds (#2203) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 04a3638b | fix(ci): pass CUDA compiler arg for Windows detection (#2202) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 2bfd80c1 | fix(ci): pass CUDA unsupported compiler flag during detection (#2201) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 9cf0ce7c | chore: bump version to 0.3.22 (#2200) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| d2113a14 | feat(ci): re-enable Windows CUDA wheels (#2198) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 587d94a8 | feat: Update llama.cpp to ggerganov/llama.cpp@63d93d173 (#2197) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| c6dc9055 | fix(docs): update mkdocstrings inventories config (#2195) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| d2bcbac4 | fix(ci): Scope CPU release wheel selectors by OS (#2194) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 195cc59a | fix(ci): Repair py3 CPU release wheels (#2193) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| c8075d1d | chore: bump version to 0.3.21 (#2192) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 511b3f41 | fix(ci): Build one arm64 py3 release wheel (#2191) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| d87bf088 | feat: Update llama.cpp to ggerganov/llama.cpp@f53577432 (#2189) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 1b1a320d | feat: Update llama.cpp to ggerganov/llama.cpp@227ed28e1 (#2182) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 1bcc5bc7 | feat: Update llama.cpp to ggerganov/llama.cpp@3bd9aa1f9 (#2176) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 02d6beed | chore: bump version to 0.3.20 (#2171) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 08e088cd | fix(misc): replace deprecated llama.cpp references (#2170) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 100b275c | feat: Update llama.cpp to ggerganov/llama.cpp@f49e9178767d557a522618b16ce8694... |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 7257ba95 | feat(server): add model-load chat_template_kwargs (#2168) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 7613aca6 | ci: publish release wheels as py3-none (#2166) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| fcd932a1 | fix(ci): publish distinct manylinux and musllinux cpu wheels (#2165) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| f54421ba | Bump version to 0.3.19 (#2162) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| c670222c | feat: Update llama.cpp to ggerganov/llama.cpp@c0159f9c1f874da15e94f371d136f59... |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| ac59e5a5 | fix: handle embedding models without KV memory (#2160) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 5f9c231c | fix(ci): reduce CUDA binary wheel size only including cubins for current arch... |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| d6f46a50 | chore: bump version (#2157) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 7b38c312 |
feat: expose attention_type parameter in Llama.__init__ (#2143)
Co-authored-by: Victor Biederbeck <v****r@m****z>, abetlen <a****n@g****m> |
Victor Biederbeck <j****s@j****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| ccc6bc04 | fix(ci): docker build workflow (#2156) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 909ebf12 | fix(ci): cuda wheel workflow (#2155) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| f0391c5e | fix(ci): release wheel workflow (#2154) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| a6b18072 | chore: Bump version (#2153) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 11e7a55a | fix: Qwen 3.5 support (#2152) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| e1f8ac0e |
ci: add riscv64 wheel builds to release workflow (#2139)
Co-authored-by: abetlen <a****n@g****m> |
Bruno Verachten <g****r@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 18aa31ef | feat: Update llama.cpp to ggerganov/llama.cpp@49bfddeca18e62fa3d39114a23e9fcb... |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| a9b4a067 | misc: Add Ruff formatting (#2148) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 9f661ff2 | fix(ci): Fix macos tests, support both Intel and Apple Silicon testing (#2150) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| ca3b00a2 | fix(ci): Rename `huggingface-cli` to `hf` (#2149) |
Andrei <a****n@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| c37132ba | chore: Bump version | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| ce6fd8bb | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| dfc9bf50 | chore: Bump version | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 30ddd56e | fix: rename op_offloat to op_offload in llama.py (#2046) |
sergey21000 <6****0@u****m>
Committed by: GitHub <n****y@g****m> |
about 1 year ago | |
| af637928 | feat: Add gpt-oss chat format support through strftime_now in chat format by ... | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 68e89e86 | misc: Add Python 3.13 classifier tag | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| d12ca479 | misc: Update pypi downloads badge | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 4f260288 | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| e1af05f4 | chore: Bump version | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 95292e36 | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| d9749cb9 | chore: Bump version | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| c8579d76 | fix: Better chat format for Qwen2.5-VL (#2040) |
Alcoft <1****O@u****m>
Committed by: GitHub <n****y@g****m> |
about 1 year ago | |
| a99fd214 | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| cce48873 | fix(ci): Fix macos cpu builds | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 8866fbd7 | chore: Bump version | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 98fda8cd | fix(ci): Temporarily disable windows cuda wheels | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| b39e9d45 | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 7011bc19 | fix(ci): Update docker runner | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 82ad829c | fix(ci): update runners for cpu builds | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 1580839f | chore: Bump version | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 11d28df7 | fix(ci): Remove macos-13 builds to fix cross compilation error | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 083fcf65 | fix(ci): Add git to package list | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| ae54cde8 | fix(ci): Update cuda build action to use ubuntu 22.04 | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 9e5a4eaa | fix: Update reference to in Llama.embed. Closes #2037 | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 9770b840 | chore: Bump version | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 6f3f0bf4 | docs: Add Qwen2.5-VL to README | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 07a979f9 | fix: Use num_threads from llama model for mtmd | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| cd548bd0 | feat: Add support for new mtmd api, add Qwen2.5-VL chat handler | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 0dec7883 | fix: Fix missing deprecated symbols on windows with missing LLAMA_API prefix ... | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 5a635f49 | fix(minor): Fix type hint for older versions of python | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 51dce741 | misc: Fix support for new parameters, deprecate rpc_servers parameter | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| 0d475d7b | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | about 1 year ago | |
| b1d23df0 | hotfix: Disable curl support | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| cb2edb9c | chore: Bump version | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| 4c6514d3 | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| 99f2ebfd | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| 37eb5f0a | chore: Bump version | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| e232fae1 | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| 344c106a | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| 710e19a8 | chore: Bump version | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| 4442ff8e |
fix: error showing time spent in llama perf context print (#1898)
Co-authored-by: Andrei <a****n@g****m> |
Shaka Huang <s****a@g****m>
Committed by: GitHub <n****y@g****m> |
over 1 year ago | |
| 14879c78 | fix(ci): Fix the CUDA workflow (#1894) |
oobabooga <o****4@g****m>
Committed by: GitHub <n****y@g****m> |
over 1 year ago | |
| 0b89fe48 | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| 80be68ac | feat: Update llama.cpp | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| 0580cf27 | chore: Bump version | Andrei Betlen <a****n@g****m> | over 1 year ago | |
| e8f14cef |
fix: streaming resource lock (#1879)
Co-authored-by: Andrei <a****n@g****m> |
Graeme Power <g****r@t****e>
Committed by: GitHub <n****y@g****m> |
over 1 year ago |