Hugging Face's logo

tensorblock
/
NuminaMath-7B-TIR-GGUF

Text Generation
GGUF
alignment-handbook
Generated from Trainer
TensorBlock
GGUF
Inference Endpoints
conversational
Model card Files Community
morriszms commited on
Commit
eea133a
•
1 Parent(s): f91f094

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -12
README.md CHANGED
@@ -62,8 +62,16 @@ This repo contains GGUF format model files for [AI-MO/NuminaMath-7B-TIR](https:/
62
 
63
  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).
64
 
 
 
 
 
 
 
 
65
  ## Prompt template
66
 
 
67
  ```
68
  ### Problem: {prompt}
69
  ### Solution:
@@ -73,18 +81,18 @@ The files were quantized using machines provided by [TensorBlock](https://tensor
73
 
74
  | Filename | Quant type | File Size | Description |
75
  | -------- | ---------- | --------- | ----------- |
76
- | [NuminaMath-7B-TIR-Q2_K.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q2_K.gguf) | Q2_K | 2.532 GB | smallest, significant quality loss - not recommended for most purposes |
77
- | [NuminaMath-7B-TIR-Q3_K_S.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q3_K_S.gguf) | Q3_K_S | 2.923 GB | very small, high quality loss |
78
- | [NuminaMath-7B-TIR-Q3_K_M.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q3_K_M.gguf) | Q3_K_M | 3.223 GB | very small, high quality loss |
79
- | [NuminaMath-7B-TIR-Q3_K_L.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q3_K_L.gguf) | Q3_K_L | 3.489 GB | small, substantial quality loss |
80
- | [NuminaMath-7B-TIR-Q4_0.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q4_0.gguf) | Q4_0 | 3.725 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
81
- | [NuminaMath-7B-TIR-Q4_K_S.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q4_K_S.gguf) | Q4_K_S | 3.749 GB | small, greater quality loss |
82
- | [NuminaMath-7B-TIR-Q4_K_M.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q4_K_M.gguf) | Q4_K_M | 3.933 GB | medium, balanced quality - recommended |
83
- | [NuminaMath-7B-TIR-Q5_0.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q5_0.gguf) | Q5_0 | 4.481 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
84
- | [NuminaMath-7B-TIR-Q5_K_S.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q5_K_S.gguf) | Q5_K_S | 4.481 GB | large, low quality loss - recommended |
85
- | [NuminaMath-7B-TIR-Q5_K_M.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q5_K_M.gguf) | Q5_K_M | 4.588 GB | large, very low quality loss - recommended |
86
- | [NuminaMath-7B-TIR-Q6_K.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q6_K.gguf) | Q6_K | 5.284 GB | very large, extremely low quality loss |
87
- | [NuminaMath-7B-TIR-Q8_0.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/tree/main/NuminaMath-7B-TIR-Q8_0.gguf) | Q8_0 | 6.842 GB | very large, extremely low quality loss - not recommended |
88
 
89
 
90
  ## Downloading instruction
 
62
 
63
  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).
64
 
65
+
66
+ <div style="text-align: left; margin: 20px 0;">
67
+ <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;">
68
+ Run them on the TensorBlock client using your local machine ↗
69
+ </a>
70
+ </div>
71
+
72
  ## Prompt template
73
 
74
+
75
  ```
76
  ### Problem: {prompt}
77
  ### Solution:
 
81
 
82
  | Filename | Quant type | File Size | Description |
83
  | -------- | ---------- | --------- | ----------- |
84
+ | [NuminaMath-7B-TIR-Q2_K.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q2_K.gguf) | Q2_K | 2.532 GB | smallest, significant quality loss - not recommended for most purposes |
85
+ | [NuminaMath-7B-TIR-Q3_K_S.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q3_K_S.gguf) | Q3_K_S | 2.923 GB | very small, high quality loss |
86
+ | [NuminaMath-7B-TIR-Q3_K_M.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q3_K_M.gguf) | Q3_K_M | 3.223 GB | very small, high quality loss |
87
+ | [NuminaMath-7B-TIR-Q3_K_L.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q3_K_L.gguf) | Q3_K_L | 3.489 GB | small, substantial quality loss |
88
+ | [NuminaMath-7B-TIR-Q4_0.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q4_0.gguf) | Q4_0 | 3.725 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
89
+ | [NuminaMath-7B-TIR-Q4_K_S.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q4_K_S.gguf) | Q4_K_S | 3.749 GB | small, greater quality loss |
90
+ | [NuminaMath-7B-TIR-Q4_K_M.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q4_K_M.gguf) | Q4_K_M | 3.933 GB | medium, balanced quality - recommended |
91
+ | [NuminaMath-7B-TIR-Q5_0.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q5_0.gguf) | Q5_0 | 4.481 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
92
+ | [NuminaMath-7B-TIR-Q5_K_S.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q5_K_S.gguf) | Q5_K_S | 4.481 GB | large, low quality loss - recommended |
93
+ | [NuminaMath-7B-TIR-Q5_K_M.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q5_K_M.gguf) | Q5_K_M | 4.588 GB | large, very low quality loss - recommended |
94
+ | [NuminaMath-7B-TIR-Q6_K.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q6_K.gguf) | Q6_K | 5.284 GB | very large, extremely low quality loss |
95
+ | [NuminaMath-7B-TIR-Q8_0.gguf](https://huggingface.co/tensorblock/NuminaMath-7B-TIR-GGUF/blob/main/NuminaMath-7B-TIR-Q8_0.gguf) | Q8_0 | 6.842 GB | very large, extremely low quality loss - not recommended |
96
 
97
 
98
  ## Downloading instruction