Skip to content

Commit 19592f4

Browse files
committed
build: Moved ibm-fms to optional dependecy aiu
Signed-off-by: Brandon Groth <[email protected]>
1 parent b9dd81e commit 19592f4

File tree

2 files changed

+3
-1
lines changed

2 files changed

+3
-1
lines changed

README.md

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -104,6 +104,7 @@ The following optional dependencies are available:
104104
- `gptq`: `GPTQModel` package for W4A16 quantization
105105
- `mx`: `microxcaling` package for MX quantization
106106
- `opt`: Shortcut for `fp8`, `gptq`, and `mx` installs
107+
- `aiu`: `ibm-fms` package for AIU model deployment
107108
- `torchvision`: `torch` package for image recognition training and inference
108109
- `triton`: `triton` package for matrix multiplication kernels
109110
- `examples`: Dependencies needed for examples

pyproject.toml

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -30,7 +30,7 @@ dependencies = [
3030
"datasets>=3.0.0,<4.0",
3131
"pandas",
3232
"safetensors",
33-
"ibm-fms>=0.0.8",
33+
3434
"pkginfo>1.10",
3535
]
3636

@@ -40,6 +40,7 @@ fp8 = ["llmcompressor"]
4040
gptq = ["Cython", "gptqmodel>=1.7.3"]
4141
mx = ["microxcaling>=1.1"]
4242
opt = ["fms-model-optimizer[fp8, gptq, mx]"]
43+
aiu = ["ibm-fms>=0.0.8"]
4344
torchvision = ["torchvision>=0.17"]
4445
flash-attn = ["flash-attn>=2.5.3,<3.0"]
4546
triton = ["triton>=3.0,<3.4"]

0 commit comments

Comments
 (0)