llmr 0.2.0

Reach language models across providers, with capabilities you can read before you ask and usage you can trust.
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

# The Gemini models this release knows about.
#
# A starting point, not a source of truth. Every row was read off a vendor model page on the
# date it carries. Supply your own Registry when you need one that is current:
# `Registry::stale` and `Registry::unlisted` compare this table against what the API says it
# serves.
#
# `streaming` is true on every row and comes from the API reference rather than the model
# pages: `models.streamGenerateContent` is a method on the generate content API, not a
# per-model feature. The model pages do not list it, which is why this note is here.
#
# The token limits are the vendor's own figures and are not round numbers: 1,048,576 in and
# 65,536 out. Rounding them to a million and 64k would be a smaller context window than the
# model has, or a larger one, and neither is what was published.

provider = "gemini"
reach    = "FirstPartyApi"

[[model]]
id                = "gemini-3.7-flash"
context_window    = 1048576
max_output        = 65536
tools             = true
structured_output = true
prompt_caching    = true
thinking          = true
images            = true
streaming         = true
source            = "vendor model page; streamGenerateContent in the API reference"
verified_at       = "2026-08-31"

[[model]]
id                = "gemini-3.5-flash"
context_window    = 1048576
max_output        = 65536
tools             = true
structured_output = true
prompt_caching    = true
thinking          = true
images            = true
streaming         = true
source            = "vendor model page; streamGenerateContent in the API reference"
verified_at       = "2026-08-31"

[[model]]
id                = "gemini-3.5-flash-lite"
context_window    = 1048576
max_output        = 65536
tools             = true
structured_output = true
prompt_caching    = true
thinking          = true
images            = true
streaming         = true
source            = "vendor model page; streamGenerateContent in the API reference"
verified_at       = "2026-08-31"

[[model]]
id                = "gemini-2.5-flash"
context_window    = 1048576
max_output        = 65536
tools             = true
structured_output = true
prompt_caching    = true
thinking          = true
images            = true
streaming         = true
source            = "vendor model page; streamGenerateContent in the API reference"
verified_at       = "2026-08-31"