Transformers
GGUF
MLX
English
Chinese
unsloth
fine tune
heretic
abliterated
uncensored
creative
creative writing
fiction writing
plot generation
sub-plot generation
story generation
scene continue
storytelling
fiction story
science fiction
romance
all genres
story
writing
vivid prosing
vivid writing
fiction
roleplaying
bfloat16
all use cases
Deckard(qx)
mxfp8
mxfp4
Merge
mergekit
conversational
auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -73,6 +73,18 @@ more details, including on how to concatenate multi-part files.
|
|
| 73 |
|:-----|:-----|--------:|:------|
|
| 74 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.mmproj-Q8_0.gguf) | mmproj-Q8_0 | 0.5 | multi-modal supplement |
|
| 75 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.mmproj-f16.gguf) | mmproj-f16 | 0.8 | multi-modal supplement |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
|
| 77 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 78 |
types (lower is better):
|
|
|
|
| 73 |
|:-----|:-----|--------:|:------|
|
| 74 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.mmproj-Q8_0.gguf) | mmproj-Q8_0 | 0.5 | multi-modal supplement |
|
| 75 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.mmproj-f16.gguf) | mmproj-f16 | 0.8 | multi-modal supplement |
|
| 76 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q2_K.gguf) | Q2_K | 2.0 | |
|
| 77 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q3_K_S.gguf) | Q3_K_S | 2.2 | |
|
| 78 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q3_K_M.gguf) | Q3_K_M | 2.4 | lower quality |
|
| 79 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q3_K_L.gguf) | Q3_K_L | 2.5 | |
|
| 80 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.IQ4_XS.gguf) | IQ4_XS | 2.6 | |
|
| 81 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q4_K_S.gguf) | Q4_K_S | 2.7 | fast, recommended |
|
| 82 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q4_K_M.gguf) | Q4_K_M | 2.8 | fast, recommended |
|
| 83 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q5_K_S.gguf) | Q5_K_S | 3.1 | |
|
| 84 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q5_K_M.gguf) | Q5_K_M | 3.2 | |
|
| 85 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q6_K.gguf) | Q6_K | 3.6 | very good quality |
|
| 86 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.Q8_0.gguf) | Q8_0 | 4.6 | fast, best quality |
|
| 87 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3.5-4B-Element-GGUF/resolve/main/Qwen3.5-4B-Element.f16.gguf) | f16 | 8.5 | 16 bpw, overkill |
|
| 88 |
|
| 89 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 90 |
types (lower is better):
|