Hugging Face's logo

mradermacher
/
Phind-CodeLlama-34B-Python-v1-GGUF

Transformers
GGUF
English
code llama
Inference Endpoints
Model card Files Community
mradermacher commited on
Commit
a06abbf
•
1 Parent(s): eb7d299

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -0
README.md CHANGED
@@ -32,8 +32,18 @@ more details, including on how to concatenate multi-part files.
32
  | Link | Type | Size/GB | Notes |
33
  |:-----|:-----|--------:|:------|
34
  | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q2_K.gguf) | Q2_K | 12.6 | |
 
 
35
  | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.IQ3_S.gguf) | IQ3_S | 14.8 | beats Q3_K* |
 
 
 
36
  | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q4_K_S.gguf) | Q4_K_S | 19.3 | fast, recommended |
 
 
 
 
 
37
 
38
  Here is a handy graph by ikawrakow comparing some lower-quality quant
39
  types (lower is better):
 
32
  | Link | Type | Size/GB | Notes |
33
  |:-----|:-----|--------:|:------|
34
  | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q2_K.gguf) | Q2_K | 12.6 | |
35
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.IQ3_XS.gguf) | IQ3_XS | 14.0 | |
36
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q3_K_S.gguf) | Q3_K_S | 14.7 | |
37
  | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.IQ3_S.gguf) | IQ3_S | 14.8 | beats Q3_K* |
38
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.IQ3_M.gguf) | IQ3_M | 15.3 | |
39
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q3_K_M.gguf) | Q3_K_M | 16.4 | lower quality |
40
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q3_K_L.gguf) | Q3_K_L | 17.9 | |
41
  | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q4_K_S.gguf) | Q4_K_S | 19.3 | fast, recommended |
42
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q4_K_M.gguf) | Q4_K_M | 20.3 | fast, recommended |
43
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q5_K_S.gguf) | Q5_K_S | 23.3 | |
44
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q5_K_M.gguf) | Q5_K_M | 23.9 | |
45
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q6_K.gguf) | Q6_K | 27.8 | very good quality |
46
+ | [GGUF](https://huggingface.co/mradermacher/Phind-CodeLlama-34B-Python-v1-GGUF/resolve/main/Phind-CodeLlama-34B-Python-v1.Q8_0.gguf) | Q8_0 | 36.0 | fast, best quality |
47
 
48
  Here is a handy graph by ikawrakow comparing some lower-quality quant
49
  types (lower is better):