genai-prices/prices/providers/modal.yml at main · pydantic/genai-prices · 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
# yaml-language-server: $schema=.schema.json
name: Modal
id: modal
pricing_urls:
  - https://modal.com/library
api_pattern: 'https://[^/]+\.modal\.(?:run|direct)(?:/|$)'
provider_match:
  contains: modal

# Modal Shared Endpoints use the OpenAI-compatible chat completions and responses APIs.
extractors:
  - api_flavor: chat
    root: usage
    mappings:
      - path: prompt_tokens
        dest: input_tokens
      - path: [prompt_tokens_details, cached_tokens]
        dest: cache_read_tokens
        required: false
      - path: [prompt_tokens_details, cache_write_tokens]
        dest: cache_write_tokens
        required: false
      - path: reasoning_tokens
        dest: output_reasoning_tokens
        required: false
      - path: completion_tokens
        dest: output_tokens

  - api_flavor: responses
    root: usage
    mappings:
      - path: input_tokens
        dest: input_tokens
      - path: [input_tokens_details, cached_tokens]
        dest: cache_read_tokens
        required: false
      - path: [input_tokens_details, cache_write_tokens]
        dest: cache_write_tokens
        required: false
      - path: [output_tokens_details, reasoning_tokens]
        dest: output_reasoning_tokens
        required: false
      - path: output_tokens
        dest: output_tokens

models:
  - id: moonshotai/Kimi-K3
    name: Kimi K3
    description: >-
      Moonshot AI's 2.8-trillion-parameter multimodal reasoning model, served by Modal as a Shared Endpoint.
    match:
      equals: moonshotai/Kimi-K3
    context_window: 1048576
    prices_checked: 2026-08-12
    price_comments: >-
      Modal Shared Endpoint pricing. Reasoning tokens use the same $15/MTok rate as completion tokens.
      See https://modal.com/library/moonshot/kimi-k3.
    prices:
      input_mtok: 3.00
      cache_read_mtok: 0.30
      output_mtok: 15.00

  - id: thinkingmachines/Inkling-NVFP4
    name: Inkling NVFP4
    description: >-
      Thinking Machines Lab's reasoning model, served by Modal as a Shared Endpoint.
    match:
      equals: thinkingmachines/Inkling-NVFP4
    prices_checked: 2026-08-12
    price_comments: >-
      Modal Shared Endpoint dashboard pricing. Reasoning tokens use the same $5/MTok rate as completion tokens.
    prices:
      input_mtok: 1.20
      cache_read_mtok: 0.27
      output_mtok: 5.00