mradermacher commited on
Commit
d187ca6
·
verified ·
1 Parent(s): 1d69fd1

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -0
README.md CHANGED
@@ -43,12 +43,15 @@ more details, including on how to concatenate multi-part files.
43
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q3_K_S.gguf) | Q3_K_S | 1.6 | |
44
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q3_K_M.gguf) | Q3_K_M | 1.7 | lower quality |
45
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q3_K_L.gguf) | Q3_K_L | 1.9 | |
 
 
46
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q4_K_S.gguf) | Q4_K_S | 2.0 | fast, recommended |
47
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q4_K_M.gguf) | Q4_K_M | 2.1 | fast, recommended |
48
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q5_K_S.gguf) | Q5_K_S | 2.4 | |
49
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q5_K_M.gguf) | Q5_K_M | 2.5 | |
50
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q6_K.gguf) | Q6_K | 2.9 | very good quality |
51
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q8_0.gguf) | Q8_0 | 3.7 | fast, best quality |
 
52
 
53
  Here is a handy graph by ikawrakow comparing some lower-quality quant
54
  types (lower is better):
 
43
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q3_K_S.gguf) | Q3_K_S | 1.6 | |
44
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q3_K_M.gguf) | Q3_K_M | 1.7 | lower quality |
45
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q3_K_L.gguf) | Q3_K_L | 1.9 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.IQ4_XS.gguf) | IQ4_XS | 1.9 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q4_0_4_4.gguf) | Q4_0_4_4 | 2.0 | fast on arm, low quality |
48
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q4_K_S.gguf) | Q4_K_S | 2.0 | fast, recommended |
49
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q4_K_M.gguf) | Q4_K_M | 2.1 | fast, recommended |
50
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q5_K_S.gguf) | Q5_K_S | 2.4 | |
51
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q5_K_M.gguf) | Q5_K_M | 2.5 | |
52
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q6_K.gguf) | Q6_K | 2.9 | very good quality |
53
  | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.Q8_0.gguf) | Q8_0 | 3.7 | fast, best quality |
54
+ | [GGUF](https://huggingface.co/mradermacher/Tiny-Llama-Llama-Dolphin-laser-1b-moe-GGUF/resolve/main/Tiny-Llama-Llama-Dolphin-laser-1b-moe.f16.gguf) | f16 | 6.9 | 16 bpw, overkill |
55
 
56
  Here is a handy graph by ikawrakow comparing some lower-quality quant
57
  types (lower is better):