Transformers
GGUF
English
esper
esper-3.1
esper-3
valiant
valiant-labs
qwen
qwen-3
qwen-3-4b
qwen3-4b-thinking-2507
4b
reasoning
code
code-instruct
python
javascript
dev-ops
jenkins
terraform
ansible
docker
kubernetes
helm
grafana
prometheus
shell
bash
azure
aws
gcp
cloud
scripting
powershell
problem-solving
architect
engineer
developer
creative
analytical
expert
rationality
conversational
chat
instruct
imatrix
auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -88,6 +88,30 @@ more details, including on how to concatenate multi-part files.
|
|
| 88 |
| Link | Type | Size/GB | Notes |
|
| 89 |
|:-----|:-----|--------:|:------|
|
| 90 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 91 |
|
| 92 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 93 |
types (lower is better):
|
|
|
|
| 88 |
| Link | Type | Size/GB | Notes |
|
| 89 |
|:-----|:-----|--------:|:------|
|
| 90 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
|
| 91 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ1_S.gguf) | i1-IQ1_S | 1.2 | for the desperate |
|
| 92 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ1_M.gguf) | i1-IQ1_M | 1.2 | mostly desperate |
|
| 93 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.3 | |
|
| 94 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ2_XS.gguf) | i1-IQ2_XS | 1.5 | |
|
| 95 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ2_S.gguf) | i1-IQ2_S | 1.5 | |
|
| 96 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ2_M.gguf) | i1-IQ2_M | 1.6 | |
|
| 97 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q2_K_S.gguf) | i1-Q2_K_S | 1.7 | very low quality |
|
| 98 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q2_K.gguf) | i1-Q2_K | 1.8 | IQ3_XXS probably better |
|
| 99 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 1.8 | lower quality |
|
| 100 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ3_XS.gguf) | i1-IQ3_XS | 1.9 | |
|
| 101 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q3_K_S.gguf) | i1-Q3_K_S | 2.0 | IQ3_XS probably better |
|
| 102 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ3_S.gguf) | i1-IQ3_S | 2.0 | beats Q3_K* |
|
| 103 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ3_M.gguf) | i1-IQ3_M | 2.1 | |
|
| 104 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q3_K_M.gguf) | i1-Q3_K_M | 2.2 | IQ3_S probably better |
|
| 105 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q3_K_L.gguf) | i1-Q3_K_L | 2.3 | IQ3_M probably better |
|
| 106 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ4_XS.gguf) | i1-IQ4_XS | 2.4 | |
|
| 107 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q4_0.gguf) | i1-Q4_0 | 2.5 | fast, low quality |
|
| 108 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-IQ4_NL.gguf) | i1-IQ4_NL | 2.5 | prefer IQ4_XS |
|
| 109 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q4_K_S.gguf) | i1-Q4_K_S | 2.5 | optimal size/speed/quality |
|
| 110 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q4_K_M.gguf) | i1-Q4_K_M | 2.6 | fast, recommended |
|
| 111 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q4_1.gguf) | i1-Q4_1 | 2.7 | |
|
| 112 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q5_K_S.gguf) | i1-Q5_K_S | 2.9 | |
|
| 113 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q5_K_M.gguf) | i1-Q5_K_M | 3.0 | |
|
| 114 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Thinking-2507-Esper3.1-i1-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Esper3.1.i1-Q6_K.gguf) | i1-Q6_K | 3.4 | practically like static Q6_K |
|
| 115 |
|
| 116 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 117 |
types (lower is better):
|