bluryar commited on
Commit
45b3628
·
verified ·
1 Parent(s): e0607e5

Update X-Voice GGUF model card

Browse files
Files changed (1) hide show
  1. README.md +17 -0
README.md CHANGED
@@ -38,6 +38,23 @@ The quantized files use a conservative X-Voice policy: large GGML `mul_mat`
38
  matrix weights are converted, while conv, norm, bias, embedding, positional, and
39
  small tensors remain in their source type.
40
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
41
  ## Quick Start
42
 
43
  ```bash
 
38
  matrix weights are converted, while conv, norm, bias, embedding, positional, and
39
  small tensors remain in their source type.
40
 
41
+ ## Quantization Benchmark
42
+
43
+ Single-run local benchmark on an RTX 4060 Ti with `NVIDIA_TF32_OVERRIDE=0`,
44
+ `--preset product`, `cfg_nonlayered`, 32 sampler steps, the default zh sample
45
+ text, and `/root/code/ggbond/models/test.wav`:
46
+
47
+ | model | size | wall | sampler | load | mel max abs vs f32 | mel mean abs vs f32 | wav max abs vs f32 | wav mean abs vs f32 |
48
+ |---|---:|---:|---:|---:|---:|---:|---:|---:|
49
+ | f32 | 1544.1 MiB | 10.74s | 9.222s | 0.515s | 0 | 0 | 0 | 0 |
50
+ | f16 | 1060.5 MiB | 8.89s | 7.316s | 0.513s | 1.40578 | 0.01232 | 0.89212 | 0.01086 |
51
+ | q8_0 | 847.4 MiB | 8.52s | 7.268s | 0.303s | 0.78026 | 0.02397 | 1.01920 | 0.01708 |
52
+ | q6_k | 790.0 MiB | 8.66s | 7.407s | 0.301s | 0.89031 | 0.04256 | 0.68326 | 0.02387 |
53
+ | q4_k | 728.8 MiB | 8.61s | 7.277s | 0.319s | 5.72519 | 0.28411 | 1.22644 | 0.07608 |
54
+
55
+ Interpretation: f16/q8_0/q6_k are the practical release candidates. q4_k is
56
+ size-first and should be auditioned carefully before use as a quality default.
57
+
58
  ## Quick Start
59
 
60
  ```bash