philipp-zettl commited on
Commit
3a2e5b9
·
verified ·
1 Parent(s): 3169511

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +73 -0
README.md ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ datasets:
4
+ - ds4sd/SynthCodeNet
5
+ - ds4sd/SynthFormulaNet
6
+ - ds4sd/SynthChartNet
7
+ - HuggingFaceM4/DoclingMatix
8
+ tags:
9
+ - text-generation
10
+ - documents
11
+ - code
12
+ - formula
13
+ - chart
14
+ - ocr
15
+ - layout
16
+ - table
17
+ - document-parse
18
+ - docling
19
+ - granite
20
+ - extraction
21
+ - math
22
+ - llama-cpp
23
+ - gguf-my-repo
24
+ language:
25
+ - en
26
+ pipeline_tag: image-text-to-text
27
+ library_name: transformers
28
+ base_model: ibm-granite/granite-docling-258M
29
+ ---
30
+
31
+ # philipp-zettl/granite-docling-258M-Q8_0-GGUF
32
+ This model was converted to GGUF format from [`ibm-granite/granite-docling-258M`](https://huggingface.co/ibm-granite/granite-docling-258M) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
33
+ Refer to the [original model card](https://huggingface.co/ibm-granite/granite-docling-258M) for more details on the model.
34
+
35
+ ## Use with llama.cpp
36
+ Install llama.cpp through brew (works on Mac and Linux)
37
+
38
+ ```bash
39
+ brew install llama.cpp
40
+
41
+ ```
42
+ Invoke the llama.cpp server or the CLI.
43
+
44
+ ### CLI:
45
+ ```bash
46
+ llama-cli --hf-repo philipp-zettl/granite-docling-258M-Q8_0-GGUF --hf-file granite-docling-258m-q8_0.gguf -p "The meaning to life and the universe is"
47
+ ```
48
+
49
+ ### Server:
50
+ ```bash
51
+ llama-server --hf-repo philipp-zettl/granite-docling-258M-Q8_0-GGUF --hf-file granite-docling-258m-q8_0.gguf -c 2048
52
+ ```
53
+
54
+ Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
55
+
56
+ Step 1: Clone llama.cpp from GitHub.
57
+ ```
58
+ git clone https://github.com/ggerganov/llama.cpp
59
+ ```
60
+
61
+ Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
62
+ ```
63
+ cd llama.cpp && LLAMA_CURL=1 make
64
+ ```
65
+
66
+ Step 3: Run inference through the main binary.
67
+ ```
68
+ ./llama-cli --hf-repo philipp-zettl/granite-docling-258M-Q8_0-GGUF --hf-file granite-docling-258m-q8_0.gguf -p "The meaning to life and the universe is"
69
+ ```
70
+ or
71
+ ```
72
+ ./llama-server --hf-repo philipp-zettl/granite-docling-258M-Q8_0-GGUF --hf-file granite-docling-258m-q8_0.gguf -c 2048
73
+ ```