ravel/benchmarks/results/t3code-ravel-v0.2.4-vs-graphify-0.9.12-2026-07-12.json at main · 12vault/ravel · 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
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
{
  "version": 1,
  "measuredAt": "2026-07-12",
  "metric": "compact_context_payload_estimated_tokens",
  "tokenEstimate": "ceil(UTF-8 bytes / 3)",
  "tokenBudget": 800,
  "cases": 10,
  "repository": "t3tools/t3code",
  "revision": "c1ec1915fc16f3dc1ec5d47d9a97f6210a574526",
  "machine": {
    "cpu": "Apple M1 Pro",
    "os": "darwin",
    "architecture": "arm64"
  },
  "ravelVersion": "ravel v0.2.4",
  "graphifyVersion": "graphify 0.9.12",
  "ravelGraphNodes": 296964,
  "ravelGraphEdges": 367445,
  "graphifyGraphNodes": 49517,
  "graphifyGraphEdges": 104160,
  "build": {
    "ravelFilesAnalyzed": 5630,
    "ravelSeconds": 148.95,
    "graphifyCodeFilesScanned": 5448,
    "graphifySeconds": 90.52,
    "graphifyRawEdges": 123747,
    "graphifyZeroNodeFiles": 805,
    "graphifySkippedSQLFiles": 11
  },
  "reclustering": {
    "ravelSeconds": 13.18,
    "graphifySeconds": 19.71,
    "graphifyBackend": "networkx-louvain",
    "graphifyNetworkxVersion": "3.6.1"
  },
  "ravelEstimatedTokens": 6851,
  "graphifyEstimatedTokens": 9009,
  "limitations": "Payload size only; this does not measure answer correctness or model billing.",
  "results": [
    {
      "id": "websocket-reconnect",
      "question": "How does the web client connect to the server and recover from WebSocket disconnects?",
      "ravelBytes": 2159,
      "ravelEstimatedTokens": 720,
      "graphifyBytes": 2629,
      "graphifyEstimatedTokens": 877
    },
    {
      "id": "provider-session-lifecycle",
      "question": "Where are coding-agent provider sessions created, resumed, and terminated?",
      "ravelBytes": 2271,
      "ravelEstimatedTokens": 757,
      "graphifyBytes": 2685,
      "graphifyEstimatedTokens": 895
    },
    {
      "id": "provider-event-translation",
      "question": "How are Codex app-server events translated into the shared client protocol?",
      "ravelBytes": 2056,
      "ravelEstimatedTokens": 686,
      "graphifyBytes": 2822,
      "graphifyEstimatedTokens": 941
    },
    {
      "id": "conversation-restoration",
      "question": "How are conversations persisted and restored after the server restarts?",
      "ravelBytes": 1981,
      "ravelEstimatedTokens": 661,
      "graphifyBytes": 2749,
      "graphifyEstimatedTokens": 917
    },
    {
      "id": "git-worktree-flow",
      "question": "How does Git worktree creation and branch switching flow through the system?",
      "ravelBytes": 2186,
      "ravelEstimatedTokens": 729,
      "graphifyBytes": 2734,
      "graphifyEstimatedTokens": 912
    },
    {
      "id": "approval-flow",
      "question": "How are tool approval or permission requests sent to the UI and resolved?",
      "ravelBytes": 1841,
      "ravelEstimatedTokens": 614,
      "graphifyBytes": 2741,
      "graphifyEstimatedTokens": 914
    },
    {
      "id": "shared-client-runtime",
      "question": "How does client-runtime share behavior between the web and mobile clients?",
      "ravelBytes": 2149,
      "ravelEstimatedTokens": 717,
      "graphifyBytes": 2730,
      "graphifyEstimatedTokens": 910
    },
    {
      "id": "streaming-output",
      "question": "How does the server stream tool calls and partial assistant output to clients?",
      "ravelBytes": 1949,
      "ravelEstimatedTokens": 650,
      "graphifyBytes": 2667,
      "graphifyEstimatedTokens": 889
    },
    {
      "id": "protocol-schemas",
      "question": "Where are WebSocket request and event schemas defined and validated?",
      "ravelBytes": 2034,
      "ravelEstimatedTokens": 678,
      "graphifyBytes": 2603,
      "graphifyEstimatedTokens": 868
    },
    {
      "id": "multi-session-ui",
      "question": "How does the frontend manage multiple agent sessions and the selected conversation?",
      "ravelBytes": 1916,
      "ravelEstimatedTokens": 639,
      "graphifyBytes": 2657,
      "graphifyEstimatedTokens": 886
    }
  ]
}