ykarout commited on
Commit
c3e4e5a
·
verified ·
1 Parent(s): ffbb0ef

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +72 -0
README.md ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ datasets:
4
+ - open-r1/Mixture-of-Thoughts
5
+ language:
6
+ - en
7
+ - ar
8
+ - fr
9
+ - es
10
+ base_model: ykarout/Mixtral-8x7B-DeepSeek-R1-Distill
11
+ pipeline_tag: text2text-generation
12
+ library_name: transformers
13
+ tags:
14
+ - reasoning
15
+ - r1
16
+ - deepseek
17
+ - mixtral
18
+ - MoE
19
+ - thinking
20
+ - code
21
+ - science
22
+ - math
23
+ - llama-cpp
24
+ - gguf-my-repo
25
+ metrics:
26
+ - accuracy
27
+ new_version: ykarout/Mixtral-8x7B-DeepSeek-R1-Distill-16bit
28
+ ---
29
+
30
+ # ykarout/Mixtral-8x7B-DeepSeek-R1-Distill-Q4_K_M-GGUF
31
+ This model was converted to GGUF format from [`ykarout/Mixtral-8x7B-DeepSeek-R1-Distill`](https://huggingface.co/ykarout/Mixtral-8x7B-DeepSeek-R1-Distill) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
32
+ Refer to the [original model card](https://huggingface.co/ykarout/Mixtral-8x7B-DeepSeek-R1-Distill) for more details on the model.
33
+
34
+ ## Use with llama.cpp
35
+ Install llama.cpp through brew (works on Mac and Linux)
36
+
37
+ ```bash
38
+ brew install llama.cpp
39
+
40
+ ```
41
+ Invoke the llama.cpp server or the CLI.
42
+
43
+ ### CLI:
44
+ ```bash
45
+ llama-cli --hf-repo ykarout/Mixtral-8x7B-DeepSeek-R1-Distill-Q4_K_M-GGUF --hf-file mixtral-8x7b-deepseek-r1-distill-q4_k_m.gguf -p "The meaning to life and the universe is"
46
+ ```
47
+
48
+ ### Server:
49
+ ```bash
50
+ llama-server --hf-repo ykarout/Mixtral-8x7B-DeepSeek-R1-Distill-Q4_K_M-GGUF --hf-file mixtral-8x7b-deepseek-r1-distill-q4_k_m.gguf -c 2048
51
+ ```
52
+
53
+ 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.
54
+
55
+ Step 1: Clone llama.cpp from GitHub.
56
+ ```
57
+ git clone https://github.com/ggerganov/llama.cpp
58
+ ```
59
+
60
+ 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).
61
+ ```
62
+ cd llama.cpp && LLAMA_CURL=1 make
63
+ ```
64
+
65
+ Step 3: Run inference through the main binary.
66
+ ```
67
+ ./llama-cli --hf-repo ykarout/Mixtral-8x7B-DeepSeek-R1-Distill-Q4_K_M-GGUF --hf-file mixtral-8x7b-deepseek-r1-distill-q4_k_m.gguf -p "The meaning to life and the universe is"
68
+ ```
69
+ or
70
+ ```
71
+ ./llama-server --hf-repo ykarout/Mixtral-8x7B-DeepSeek-R1-Distill-Q4_K_M-GGUF --hf-file mixtral-8x7b-deepseek-r1-distill-q4_k_m.gguf -c 2048
72
+ ```