Logo SVG copied to clipboard
New Introducing paid plans for Manifest Cloud Learn more

Custom providers on Manifest

26.4B tokens / 30 days

48 models routed through Manifest over the last 30 days.

Bring any OpenAI-compatible endpoint — vLLM, Ollama, LiteLLM, LM Studio, or a private gateway — and Manifest routes it like any other provider, with a spending cap per agent and automatic fallback. Every custom endpoint connected to Manifest is grouped on this page as one provider.

Custom provider tokens processed on Manifest

Models in use

# Model
Auth type Tokens / last 30 days
1.MiniMax M3
API Key8.9 B
2.Other Custom Models
API Key4.5 B
3.Claude Opus 4 8
API Key3.1 B
4.Claude Sonnet 4 6
API Key1.5 B
5.Claude Sonnet 4 6
API Key1.2 B
6.Deepseek V4 Flash Free
API Key787 M
7.Mimo V2.5 Pro Free
API Key761 M
8.Agnes 2.0 Flash
API Key528 M
9.Mimo V2.5
API Key514 M
10.Glm 5.2
API Key498 M
11.North Mini Code Free
API Key431 M
12.Minimax M3
API Key400 M
13.Minimax M2.7
API Key374 M
14.Mimo V2.5 Free
API Key359 M
15.Deepseek V4 Flash
API Key352 M
16.Kimi K2.6
API Key272 M
17.Nemotron 3 Super 120b A12b
API Key240 M
18.Mimo V2.5 Pro
API Key183 M
19.Nemotron 3 Ultra Free
API Key179 M
20.Claude Opus 4 6
API Key155 M
21.Deepseek V4 Pro
API Key154 M
22.Openrouter Free
API Key153 M
23.Gpt 5.5
API Key122 M
24.Claude Sonnet 5
API Key101 M
25.Mistral Large
API Key86 M
26.Mistral Medium 3 5
API Key76 M
27.Claude Opus 4 7
API Key75 M
28.Claude Haiku 4 5 20251001
API Key74 M
29.Big Pickle
API Key48 M
30.Claude Fable 5
API Key46 M
31.Kimi K2.6
API Key42 M
32.Glm 5.2 Free
API Key42 M
33.Kimi K2.6
API Key39 M
34.Gpt 5.4 Mini
API Key31 M
35.Glm 5.1
API Key31 M
36.Gpt 5.4
API Key31 M
37.Claude Sonnet 4.5
API Key22 M
38.Claude Haiku 4.5
API Key19 M
39.Kimi K2.7 Code
API Key13 M
40.Gpt 5.3 Codex
API Key11 M
41.Gpt Oss 120b
API Key10 M
42.Glm 5.2
API Key3 M
43.Nemotron 3 Nano Omni 30b A3b Reasoning
API Key3 M
44.Kimi K2.7 Code Free
API Key2 M
45.Gpt 5.4
API Key2 M
46.Deepseek V3.2
API Key2 M
47.Zai Glm 4.7
API Key1 M
48.Gpt Oss 120b
API Key41 K

Model names are aggregated across every connected custom endpoint. Models used by only a handful of workspaces are grouped under “Other Custom Models” to keep individual configurations private.