GGUF
draft
speculative-decoding
llama-cpp
gguf-my-repo
conversational
Gapeleon commited on
Commit
5638f21
·
verified ·
1 Parent(s): 86c9657

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +70 -0
README.md ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: jukofyork/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0
3
+ datasets:
4
+ - open-thoughts/OpenThoughts-Unverified-173k
5
+ - cognitivecomputations/dolphin-r1
6
+ language:
7
+ - zho
8
+ - eng
9
+ - fra
10
+ - spa
11
+ - por
12
+ - deu
13
+ - ita
14
+ - rus
15
+ - jpn
16
+ - kor
17
+ - vie
18
+ - tha
19
+ - ara
20
+ license: apache-2.0
21
+ tags:
22
+ - draft
23
+ - speculative-decoding
24
+ - llama-cpp
25
+ - gguf-my-repo
26
+ ---
27
+
28
+ # Gapeleon/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0-Q4_K_M-GGUF
29
+ This model was converted to GGUF format from [`jukofyork/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0`](https://huggingface.co/jukofyork/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
30
+ Refer to the [original model card](https://huggingface.co/jukofyork/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0) for more details on the model.
31
+
32
+ ## Use with llama.cpp
33
+ Install llama.cpp through brew (works on Mac and Linux)
34
+
35
+ ```bash
36
+ brew install llama.cpp
37
+
38
+ ```
39
+ Invoke the llama.cpp server or the CLI.
40
+
41
+ ### CLI:
42
+ ```bash
43
+ llama-cli --hf-repo Gapeleon/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0-Q4_K_M-GGUF --hf-file deepseek-r1-0528-coder-draft-0.6b-v1.0-q4_k_m.gguf -p "The meaning to life and the universe is"
44
+ ```
45
+
46
+ ### Server:
47
+ ```bash
48
+ llama-server --hf-repo Gapeleon/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0-Q4_K_M-GGUF --hf-file deepseek-r1-0528-coder-draft-0.6b-v1.0-q4_k_m.gguf -c 2048
49
+ ```
50
+
51
+ 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.
52
+
53
+ Step 1: Clone llama.cpp from GitHub.
54
+ ```
55
+ git clone https://github.com/ggerganov/llama.cpp
56
+ ```
57
+
58
+ 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).
59
+ ```
60
+ cd llama.cpp && LLAMA_CURL=1 make
61
+ ```
62
+
63
+ Step 3: Run inference through the main binary.
64
+ ```
65
+ ./llama-cli --hf-repo Gapeleon/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0-Q4_K_M-GGUF --hf-file deepseek-r1-0528-coder-draft-0.6b-v1.0-q4_k_m.gguf -p "The meaning to life and the universe is"
66
+ ```
67
+ or
68
+ ```
69
+ ./llama-server --hf-repo Gapeleon/DeepSeek-R1-0528-CODER-DRAFT-0.6B-v1.0-Q4_K_M-GGUF --hf-file deepseek-r1-0528-coder-draft-0.6b-v1.0-q4_k_m.gguf -c 2048
70
+ ```