GraphTok/CMakePresets.json at develop · MachineLearningSystem/GraphTok · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
{
  "version": 6,
  "cmakeMinimumRequired": { "major": 3, "minor": 24, "patch": 0 },
  "configurePresets": [
    {
      "name": "_base",
      "hidden": true,
      "binaryDir": "${sourceDir}/build/${presetName}",
      "generator": "Unix Makefiles",
      "cacheVariables": {
        "CMAKE_BUILD_TYPE": "Release",
        "CMAKE_CUDA_ARCHITECTURES": "80;90"
      }
    },
    {
      "name": "all",
      "displayName": "Default vocabs (Qwen excluded)",
      "description": "GPT-2 + Llama-3 + DeepSeek-V3 + Gemma 3. Qwen is excluded until NFC normalization is GPU-resident.",
      "inherits": "_base",
      "cacheVariables": {
        "GBPE_VOCABS": "ALL",
        "GBPE_GEMMA_GPU_PRETOK": "ON"
      }
    },
    {
      "name": "gpt2",
      "displayName": "GPT-2 only",
      "description": "Smallest, fastest GPT-2-only build. Slot64-only kernel.",
      "inherits": "_base",
      "cacheVariables": { "GBPE_VOCABS": "GPT2" }
    },
    {
      "name": "llama3",
      "displayName": "Llama-3 only",
      "description": "Llama-3-only build. Slot128-only kernel.",
      "inherits": "_base",
      "cacheVariables": { "GBPE_VOCABS": "LLAMA3" }
    },
    {
      "name": "qwen25",
      "displayName": "Qwen2.5 only",
      "description": "Qwen2.5-only build. Slot128-only kernel.",
      "inherits": "_base",
      "cacheVariables": { "GBPE_VOCABS": "QWEN25" }
    },
    {
      "name": "deepseek_v3",
      "displayName": "DeepSeek-V3 only",
      "description": "DeepSeek-V3 with three-pass pre-tokenizer. Slot128-only kernel.",
      "inherits": "_base",
      "cacheVariables": { "GBPE_VOCABS": "DEEPSEEK_V3" }
    },
    {
      "name": "bytelevel",
      "displayName": "All byte-level BPE",
      "description": "GPT-2 + Llama-3 + Qwen2.5 + DeepSeek-V3, no Gemma 3.",
      "inherits": "_base",
      "cacheVariables": { "GBPE_VOCABS": "GPT2;LLAMA3;QWEN25;DEEPSEEK_V3" }
    },
    {
      "name": "gemma3",
      "displayName": "Gemma 3, GPU pre-tokenizer",
      "description": "Gemma 3 with graph-resident normalization, initial IDs, fallback, and bucketing.",
      "inherits": "_base",
      "cacheVariables": {
        "GBPE_VOCABS": "GEMMA3",
        "GBPE_GEMMA_GPU_PRETOK": "ON"
      }
    },
    {
      "name": "gemma3_cpu",
      "displayName": "Gemma 3, CPU pre-tokenizer",
      "description": "Gemma 3 comparison build with the original host SPPreTokenizer.",
      "inherits": "_base",
      "cacheVariables": {
        "GBPE_VOCABS": "GEMMA3",
        "GBPE_GEMMA_GPU_PRETOK": "OFF"
      }
    },
    {
      "name": "gemma3_gpu",
      "displayName": "Gemma 3, GPU pre-tokenizer (compatibility alias)",
      "description": "Compatibility alias for the default gemma3 GPU-pretokenizer build.",
      "inherits": "gemma3"
    }
  ],
  "buildPresets": [
    { "name": "all",         "configurePreset": "all" },
    { "name": "gpt2",        "configurePreset": "gpt2" },
    { "name": "llama3",      "configurePreset": "llama3" },
    { "name": "qwen25",      "configurePreset": "qwen25" },
    { "name": "deepseek_v3", "configurePreset": "deepseek_v3" },
    { "name": "bytelevel",   "configurePreset": "bytelevel" },
    { "name": "gemma3",      "configurePreset": "gemma3" },
    { "name": "gemma3_cpu",  "configurePreset": "gemma3_cpu" },
    { "name": "gemma3_gpu",  "configurePreset": "gemma3_gpu" }
  ]
}