sdgoij/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-03-17 16:44:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

Updated Feature matrix (markdown)

Eve
2025-05-04 18:59:18 +00:00
parent fe1b31c723
commit 288767490f
1 changed files with 6 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
Feature-matrix.md

@@ -1,10 +1,11 @@
| | **CPU (AVX/AVX2)** | **CPU (ARM NEON)** | **Metal** | **CUDA** | **ROCm** | **SYCL** | **Vulkan** | **Kompute** |
|:-----------------------:|:--------------:|:------------------:|:---------:|:----------:|:----------------:|:----------:|:----------:|:-----------:|
| **K-quants** | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ 🐢⁵ | 🚫 |
| **I-quants** | ✅ 🐢⁴ | ✅ 🐢⁴ | ✅ 🐢⁴ | ✅ | ✅ | Partial¹ | ✅ 🐢⁴ | 🚫 |
| **Parallel Multi-GPU⁶** | N/A | N/A | N/A | ✅ | ✅ | Sequential only | Sequential only | ❓ |
| **K cache quants** | ✅ | ❓ | ✅ | ✅ | ✅ | ❓ | ✅ | 🚫 |
| **MoE architecture** | ✅ | ❓ | ✅ | ✅ | ✅ | ❓ | ✅ | 🚫 |
| **I-quants** | ✅ 🐢⁴ | ✅ 🐢⁴ | ✅ 🐢⁴ | ✅ | ✅ | Partial¹ | ✅ 🐢⁴ | 🚫 |
| **Parallel Multi-GPU⁶** | N/A | N/A | N/A | ✅ | ✅ | Sequential only | Sequential only | ❓ |
| **K cache quants** | ✅ | ✅ | ✅ | ✅ | ✅ | ❓ | ✅ | 🚫 |
| **MoE architecture** | ✅ | ✅ | ✅ | ✅ | ✅ | ❓ | ✅ | 🚫 |
| **Flash Attention** | ✅ | ✅ | ✅ | ✅ | ✅ | ❓ | Partial⁷ | 🚫 |
* ✅: feature works
* 🚫: feature does not work
@@ -16,4 +17,4 @@
* ⁴: Slower than K-quants of comparable size
* ⁵: Generally the CUDA or ROCM backends are faster, though there are cases where Vulkan has faster text generation. See #10879 for benchmarks.
* ⁶: By default, all GPU backends can utilize multiple devices by running them sequentially. The CUDA code (which is also used for ROCm via HIP) also has code for running GPUs in parallel via `--split-mode row`. However, this is optimized relatively poorly and is only faster if the interconnect speed is fast vs. the speed of a single GPU.
* ⁶: Only q8_0 and iq4_nl
* ⁷: This is only implemented for coopmat2. Otherwise the Flash Attention ops will run slowly on CPU instead.