objectiveai-sdk 2.0.8

ObjectiveAI SDK, definitions, and utilities
Documentation 
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

{
  "title": "agent.openrouter.ProviderQuantization",
  "description": "Model quantization levels for provider filtering.\n\nQuantization reduces model precision to decrease memory usage and\nincrease inference speed, potentially at the cost of output quality.",
  "anyOf": [
    {
      "title": "Int4",
      "description": "4-bit integer quantization.",
      "type": "string",
      "enum": [
        "int4"
      ]
    },
    {
      "title": "Int8",
      "description": "8-bit integer quantization.",
      "type": "string",
      "enum": [
        "int8"
      ]
    },
    {
      "title": "Fp4",
      "description": "4-bit floating point quantization.",
      "type": "string",
      "enum": [
        "fp4"
      ]
    },
    {
      "title": "Fp6",
      "description": "6-bit floating point quantization.",
      "type": "string",
      "enum": [
        "fp6"
      ]
    },
    {
      "title": "Fp8",
      "description": "8-bit floating point quantization.",
      "type": "string",
      "enum": [
        "fp8"
      ]
    },
    {
      "title": "Fp16",
      "description": "16-bit floating point (half precision).",
      "type": "string",
      "enum": [
        "fp16"
      ]
    },
    {
      "title": "Bf16",
      "description": "16-bit brain floating point.",
      "type": "string",
      "enum": [
        "bf16"
      ]
    },
    {
      "title": "Fp32",
      "description": "32-bit floating point (full precision).",
      "type": "string",
      "enum": [
        "fp32"
      ]
    },
    {
      "title": "Unknown",
      "description": "Unknown quantization level.",
      "type": "string",
      "enum": [
        "unknown"
      ]
    }
  ]
}