Skip to content

Releases: ai-dock/llama.cpp-cuda

llama.cpp b6880 with CUDA

30 Oct 02:50
@github-actions github-actions
b6880
30b773d

Choose a tag to compare

View all tags
llama.cpp b6880 with CUDA

llama.cpp b6880 with CUDA Support

Pre-built binaries of llama.cpp with CUDA support for multiple CUDA versions.

Source: https://github.com/ggml-org/llama.cpp/releases/tag/b6880
Commit: b9ce94017729465895402cbcfffb51fa926c15e3

CUDA Versions

  • CUDA 12.4 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.6 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.8 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 12.9 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 13.0 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0

Architecture Reference

  • 6.1: Titan XP, Tesla P40, GTX 10xx
  • 7.0: Tesla V100
  • 7.5: Tesla T4, RTX 20xx series, Quadro RTX
  • 8.0: A100
  • 8.6: RTX 3000 series
  • 8.9: RTX 4000 series, L4, L40
  • 9.0: H100, H200
  • 10.0: B200
  • 12.0: RTX Pro series, RTX 50xx

Usage

Download the appropriate tarball for your CUDA version and extract:

tar -xzf llama.cpp-b6880-cuda-12.8.tar.gz
./llama-cli --help
Loading

llama.cpp b6869 with CUDA

29 Oct 02:54
@github-actions github-actions
b6869
30b773d

Choose a tag to compare

View all tags
llama.cpp b6869 with CUDA

llama.cpp b6869 with CUDA Support

Pre-built binaries of llama.cpp with CUDA support for multiple CUDA versions.

Source: https://github.com/ggml-org/llama.cpp/releases/tag/b6869
Commit: 851553ea6b24cb39fd5fd188b437d777cb411de8

CUDA Versions

  • CUDA 12.4 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.6 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.8 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 12.9 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 13.0 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0

Architecture Reference

  • 6.1: Titan XP, Tesla P40, GTX 10xx
  • 7.0: Tesla V100
  • 7.5: Tesla T4, RTX 20xx series, Quadro RTX
  • 8.0: A100
  • 8.6: RTX 3000 series
  • 8.9: RTX 4000 series, L4, L40
  • 9.0: H100, H200
  • 10.0: B200
  • 12.0: RTX Pro series, RTX 50xx

Usage

Download the appropriate tarball for your CUDA version and extract:

tar -xzf llama.cpp-b6869-cuda-12.8.tar.gz
./llama-cli --help
Loading

llama.cpp b6859 with CUDA

28 Oct 02:55
@github-actions github-actions
b6859
30b773d

Choose a tag to compare

View all tags
llama.cpp b6859 with CUDA

llama.cpp b6859 with CUDA Support

Pre-built binaries of llama.cpp with CUDA support for multiple CUDA versions.

Source: https://github.com/ggml-org/llama.cpp/releases/tag/b6859
Commit: c053e18a66dd95dc340aa61317877c2a41d4e3cf

CUDA Versions

  • CUDA 12.4 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.6 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.8 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 12.9 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 13.0 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0

Architecture Reference

  • 6.1: Titan XP, Tesla P40, GTX 10xx
  • 7.0: Tesla V100
  • 7.5: Tesla T4, RTX 20xx series, Quadro RTX
  • 8.0: A100
  • 8.6: RTX 3000 series
  • 8.9: RTX 4000 series, L4, L40
  • 9.0: H100, H200
  • 10.0: B200
  • 12.0: RTX Pro series, RTX 50xx

Usage

Download the appropriate tarball for your CUDA version and extract:

tar -xzf llama.cpp-b6859-cuda-12.8.tar.gz
./llama-cli --help
Loading

llama.cpp b6849 with CUDA

27 Oct 02:56
@github-actions github-actions
b6849
30b773d

Choose a tag to compare

View all tags
llama.cpp b6849 with CUDA

llama.cpp b6849 with CUDA Support

Pre-built binaries of llama.cpp with CUDA support for multiple CUDA versions.

Source: https://github.com/ggml-org/llama.cpp/releases/tag/b6849
Commit: 75d33b9302f84a5b89f82205d2bcd8def5a64e0a

CUDA Versions

  • CUDA 12.4 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.6 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.8 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 12.9 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 13.0 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0

Architecture Reference

  • 6.1: Titan XP, Tesla P40, GTX 10xx
  • 7.0: Tesla V100
  • 7.5: Tesla T4, RTX 20xx series, Quadro RTX
  • 8.0: A100
  • 8.6: RTX 3000 series
  • 8.9: RTX 4000 series, L4, L40
  • 9.0: H100, H200
  • 10.0: B200
  • 12.0: RTX Pro series, RTX 50xx

Usage

Download the appropriate tarball for your CUDA version and extract:

tar -xzf llama.cpp-b6849-cuda-12.8.tar.gz
./llama-cli --help
Loading

llama.cpp b6838 with CUDA

26 Oct 02:55
@github-actions github-actions
b6838
30b773d

Choose a tag to compare

View all tags
llama.cpp b6838 with CUDA

llama.cpp b6838 with CUDA Support

Pre-built binaries of llama.cpp with CUDA support for multiple CUDA versions.

Source: https://github.com/ggml-org/llama.cpp/releases/tag/b6838
Commit: 226f295f4dd92ad714533adc5497afed5fa88bb8

CUDA Versions

  • CUDA 12.4 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.6 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.8 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 12.9 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 13.0 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0

Architecture Reference

  • 6.1: Titan XP, Tesla P40, GTX 10xx
  • 7.0: Tesla V100
  • 7.5: Tesla T4, RTX 20xx series, Quadro RTX
  • 8.0: A100
  • 8.6: RTX 3000 series
  • 8.9: RTX 4000 series, L4, L40
  • 9.0: H100, H200
  • 10.0: B200
  • 12.0: RTX Pro series, RTX 50xx

Usage

Download the appropriate tarball for your CUDA version and extract:

tar -xzf llama.cpp-b6838-cuda-12.8.tar.gz
./llama-cli --help
Loading

llama.cpp b6834 with CUDA

25 Oct 02:43
@github-actions github-actions
b6834
30b773d

Choose a tag to compare

View all tags
llama.cpp b6834 with CUDA

llama.cpp b6834 with CUDA Support

Pre-built binaries of llama.cpp with CUDA support for multiple CUDA versions.

Source: https://github.com/ggml-org/llama.cpp/releases/tag/b6834
Commit: 55945d2ef51b93821d4b6f4a9b994393344a90db

CUDA Versions

  • CUDA 12.4 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.6 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.8 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 12.9 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 13.0 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0

Architecture Reference

  • 6.1: Titan XP, Tesla P40, GTX 10xx
  • 7.0: Tesla V100
  • 7.5: Tesla T4, RTX 20xx series, Quadro RTX
  • 8.0: A100
  • 8.6: RTX 3000 series
  • 8.9: RTX 4000 series, L4, L40
  • 9.0: H100, H200
  • 10.0: B200
  • 12.0: RTX Pro series, RTX 50xx

Usage

Download the appropriate tarball for your CUDA version and extract:

tar -xzf llama.cpp-b6834-cuda-12.8.tar.gz
./llama-cli --help
Loading

llama.cpp b6833 with CUDA

24 Oct 14:19
@github-actions github-actions
b6833
30b773d

Choose a tag to compare

View all tags
llama.cpp b6833 with CUDA

llama.cpp b6833 with CUDA Support

Pre-built binaries of llama.cpp with CUDA support for multiple CUDA versions.

Source: https://github.com/ggml-org/llama.cpp/releases/tag/b6833
Commit: 0bcb40b48c6fc6f17ba9672625e526ab2574344b

CUDA Versions

  • CUDA 12.4 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.6 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.8 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 12.9 - Architectures: 6.1, 7.0, 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 13.0 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0

Architecture Reference

  • 6.1: Titan XP, Tesla P40, GTX 10xx
  • 7.0: Tesla V100
  • 7.5: Tesla T4, RTX 20xx series, Quadro RTX
  • 8.0: A100
  • 8.6: RTX 3000 series
  • 8.9: RTX 4000 series, L4, L40
  • 9.0: H100, H200
  • 10.0: B200
  • 12.0: RTX Pro series, RTX 50xx

Usage

Download the appropriate tarball for your CUDA version and extract:

tar -xzf llama.cpp-b6833-cuda-12.8.tar.gz
./llama-cli --help
Loading

llama.cpp b6829 with CUDA

23 Oct 21:16
@github-actions github-actions
b6829
748efa4

Choose a tag to compare

View all tags
llama.cpp b6829 with CUDA

llama.cpp b6829 with CUDA Support

Pre-built binaries of llama.cpp with CUDA support for multiple CUDA versions.

Source: https://github.com/ggml-org/llama.cpp/releases/tag/b6829
Commit: 0bf47a1dbba4d36f2aff4e8c34b06210ba34e688

CUDA Versions

  • CUDA 12.4 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.6 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0
  • CUDA 12.8 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 12.9 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0
  • CUDA 13.0 - Architectures: 7.5, 8.0, 8.6, 8.9, 9.0, 10.0, 12.0

Architecture Reference

  • 7.5: Tesla T4, RTX 2000 series, Quadro RTX
  • 8.0: A100
  • 8.6: RTX 3000 series
  • 8.9: RTX 4000 series, L4, L40
  • 9.0: H100, H200
  • 10.0: B200
  • 12.0: RTX Pro series, RTX 50xx

Usage

Download the appropriate tarball for your CUDA version and extract:

tar -xzf llama.cpp-b6829-cuda-12.8.tar.gz
./llama-cli --help
Loading