Bring Your Own LLM

Use the AI model that works best for you

Your Choice, Your Control

Don't get locked into a single AI provider. Bring your own LLM and use the model that best fits your needs, budget, and requirements. Whether you need the latest cloud models or want to run everything locally, you have the freedom to choose.

Seamless Integration

Switch between different LLM providers effortlessly. Our platform supports multiple providers, so you can use different models for different tasks, compare performance, or switch providers without changing your workflows or code.

Easy Configuration for Self Hosting and Home Labs

Configure your LLM providers with a simple config file. Add your API keys, choose your models, and you're ready to go.

heyo.conf

1[server]

2

3host = "127.0.0.1"

4port = 4444

5log_level = "info"

6

7

8[llm]

9

10default_provider = "anthropic"

11default_model = "claude-sonnet-4-20250514"

12

13

14[llm.api_keys]

15

16openai = "your-openai-api-key-here"

17anthropic = "your-anthropic-api-key-here"

18exa = "your-exa-api-key-here"

19inceptionlabs = "your-inceptionlabs-api-key-here"

20mistral = "your-mistral-api-key-here"

21gemini = "your-gemini-api-key-here"

22

23

24[llm.models]

25

26openai = "gpt-4"

27anthropic = "claude-haiku-4-5-20251001"

28ollama = "qwq:latest"

29mistral = "magistral-medium-2509"

30inceptionlabs = "mercury"

31gemini = "gemini-1.5-pro"

32

33

34[mcp]

35

36default_tools = ["websearch", "webresource"]

37bearer_token = "your_auth_token_here"

38

39[[mcp.servers]]

40

41name = "heymcp"

42url = "http://localhost:8080/sse"

43enabled = true

44tools = ["websearch", "webresource"]

45

46

47[notifications]

48

49[notifications.twilio]

50

51account_sid = "your-twilio-account-sid-here"

52auth_token = "your-twilio-auth-token-here"

53from_number = "your-twilio-phone-number-here"

54

55

56[notifications.email]

57

58smtp_server = "smtp.gmail.com"

59smtp_port = 587

60username = "your-email@gmail.com"

61password = "your-app-password"

62from_email = "your-email@gmail.com"

63

64

65[auth]

66

67use_authentication = true

68jwt_secret = "your-super-secret-jwt-key-here"

69jwt_expiration = "24h"

70multitenant = true

71

72

73[storage]

74

75driver = "postgres"

76database_url = "postgresql://postgres:your-password@localhost:5444/heyo_db"

77

78

79[storage.s3]

80

81bucket = "your-s3-bucket-name"

82region = "us-east-1"

83profile = "your-aws-profile"

84

85

86[memory]

87

88summarization_provider = "anthropic"

89summarization_model = "claude-sonnet-4-20250514"

90summarization_temperature = 0.3

91summarization_max_tokens = 5000

Multiple Providers

Support for Anthropic, Mistral, OpenAI, Google Gemini, Inception, and Ollama. Use the models that work best for your specific use cases.

Local & Cloud

Run models locally with Ollama for complete privacy, or use cloud providers for more powerful models. Mix and match as needed.

Easy Switching

Change providers or models with a simple configuration change. No need to rewrite your agents or workflows.

Cost Control

Choose providers based on cost, performance, and features. Optimize your AI spend by using the right model for each task.

Supported Providers

Connect with your preferred AI provider

Anthropic

Mistral

OpenAI

Inception

Ollama

Gemini