Card: note Linux-only builds and universal builds in progress
Browse files
README.md
CHANGED
|
@@ -65,6 +65,9 @@ Mainline llama.cpp does not load `clover_moe`. This build adds the converter, th
|
|
| 65 |
|
| 66 |
## Compatibility
|
| 67 |
|
|
|
|
|
|
|
|
|
|
| 68 |
- Model: `Diffuspace/Clover-35B-A3B-Preview` (`clover_moe`). Works with the BF16 and Q8_0 GGUFs and with third-party `clover_moe` GGUFs such as the IQ4_XS. It does not load the original safetensors.
|
| 69 |
- Linux x86_64: the CPU package runs on any Linux; the CUDA package requires an NVIDIA GPU and a CUDA 12.x runtime, with archs 80/86/89/90 (A100, H100/H200, L40S, RTX 4090, A10G). Blackwell 100/120 is not in this build.
|
| 70 |
- glibc: the Linux x64 archives require `GLIBC_2.38` or newer. Ubuntu 24.04 or equivalent is the safe baseline; older distributions need their own build from the branch.
|
|
|
|
| 65 |
|
| 66 |
## Compatibility
|
| 67 |
|
| 68 |
+
> [!NOTE]
|
| 69 |
+
> **Platform availability.** The precompiled builds published today run on Linux only. Universal builds are in progress for Windows, AMD GPUs, NVIDIA GPUs and any CPU; until they ship, any other platform needs a build from the same branch.
|
| 70 |
+
|
| 71 |
- Model: `Diffuspace/Clover-35B-A3B-Preview` (`clover_moe`). Works with the BF16 and Q8_0 GGUFs and with third-party `clover_moe` GGUFs such as the IQ4_XS. It does not load the original safetensors.
|
| 72 |
- Linux x86_64: the CPU package runs on any Linux; the CUDA package requires an NVIDIA GPU and a CUDA 12.x runtime, with archs 80/86/89/90 (A100, H100/H200, L40S, RTX 4090, A10G). Blackwell 100/120 is not in this build.
|
| 73 |
- glibc: the Linux x64 archives require `GLIBC_2.38` or newer. Ubuntu 24.04 or equivalent is the safe baseline; older distributions need their own build from the branch.
|