Hugging Face's logo

tensorblock
/
falcon-11B-GGUF

GGUF
TensorBlock
GGUF
conversational
Model card Files Community
morriszms commited on
Commit
5554819
·
verified ·
1 Parent(s): 86809a7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -12
README.md CHANGED
@@ -37,8 +37,16 @@ This repo contains GGUF format model files for [tiiuae/falcon-11B](https://huggi
37
 
38
  The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
39
 
 
 
 
 
 
 
 
40
  ## Prompt template
41
 
 
42
  ```
43
  System: {system_prompt}
44
  User:
@@ -50,18 +58,18 @@ Falcon:
50
 
51
  | Filename | Quant type | File Size | Description |
52
  | -------- | ---------- | --------- | ----------- |
53
- | [falcon-11B-Q2_K.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q2_K.gguf) | Q2_K | 3.960 GB | smallest, significant quality loss - not recommended for most purposes |
54
- | [falcon-11B-Q3_K_S.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q3_K_S.gguf) | Q3_K_S | 4.604 GB | very small, high quality loss |
55
- | [falcon-11B-Q3_K_M.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q3_K_M.gguf) | Q3_K_M | 5.063 GB | very small, high quality loss |
56
- | [falcon-11B-Q3_K_L.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q3_K_L.gguf) | Q3_K_L | 5.413 GB | small, substantial quality loss |
57
- | [falcon-11B-Q4_0.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q4_0.gguf) | Q4_0 | 5.944 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
58
- | [falcon-11B-Q4_K_S.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q4_K_S.gguf) | Q4_K_S | 5.944 GB | small, greater quality loss |
59
- | [falcon-11B-Q4_K_M.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q4_K_M.gguf) | Q4_K_M | 6.379 GB | medium, balanced quality - recommended |
60
- | [falcon-11B-Q5_0.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q5_0.gguf) | Q5_0 | 7.206 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
61
- | [falcon-11B-Q5_K_S.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q5_K_S.gguf) | Q5_K_S | 7.206 GB | large, low quality loss - recommended |
62
- | [falcon-11B-Q5_K_M.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q5_K_M.gguf) | Q5_K_M | 7.641 GB | large, very low quality loss - recommended |
63
- | [falcon-11B-Q6_K.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q6_K.gguf) | Q6_K | 8.546 GB | very large, extremely low quality loss |
64
- | [falcon-11B-Q8_0.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/tree/main/falcon-11B-Q8_0.gguf) | Q8_0 | 10.990 GB | very large, extremely low quality loss - not recommended |
65
 
66
 
67
  ## Downloading instruction
 
37
 
38
  The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
39
 
40
+
41
+ <div style="text-align: left; margin: 20px 0;">
42
+ <a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
43
+ Run them on the TensorBlock client using your local machine ↗
44
+ </a>
45
+ </div>
46
+
47
  ## Prompt template
48
 
49
+
50
  ```
51
  System: {system_prompt}
52
  User:
 
58
 
59
  | Filename | Quant type | File Size | Description |
60
  | -------- | ---------- | --------- | ----------- |
61
+ | [falcon-11B-Q2_K.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q2_K.gguf) | Q2_K | 3.960 GB | smallest, significant quality loss - not recommended for most purposes |
62
+ | [falcon-11B-Q3_K_S.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q3_K_S.gguf) | Q3_K_S | 4.604 GB | very small, high quality loss |
63
+ | [falcon-11B-Q3_K_M.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q3_K_M.gguf) | Q3_K_M | 5.063 GB | very small, high quality loss |
64
+ | [falcon-11B-Q3_K_L.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q3_K_L.gguf) | Q3_K_L | 5.413 GB | small, substantial quality loss |
65
+ | [falcon-11B-Q4_0.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q4_0.gguf) | Q4_0 | 5.944 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
66
+ | [falcon-11B-Q4_K_S.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q4_K_S.gguf) | Q4_K_S | 5.944 GB | small, greater quality loss |
67
+ | [falcon-11B-Q4_K_M.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q4_K_M.gguf) | Q4_K_M | 6.379 GB | medium, balanced quality - recommended |
68
+ | [falcon-11B-Q5_0.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q5_0.gguf) | Q5_0 | 7.206 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
69
+ | [falcon-11B-Q5_K_S.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q5_K_S.gguf) | Q5_K_S | 7.206 GB | large, low quality loss - recommended |
70
+ | [falcon-11B-Q5_K_M.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q5_K_M.gguf) | Q5_K_M | 7.641 GB | large, very low quality loss - recommended |
71
+ | [falcon-11B-Q6_K.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q6_K.gguf) | Q6_K | 8.546 GB | very large, extremely low quality loss |
72
+ | [falcon-11B-Q8_0.gguf](https://huggingface.co/tensorblock/falcon-11B-GGUF/blob/main/falcon-11B-Q8_0.gguf) | Q8_0 | 10.990 GB | very large, extremely low quality loss - not recommended |
73
 
74
 
75
  ## Downloading instruction