Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

feat: added support for A21 and Amazon Titan models via bedrock api #1101

Merged
merged 8 commits into from
Sep 17, 2024
Merged

feat: added support for A21 and Amazon Titan models via bedrock api #1101

Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
ae1f3ec
feat: added support for A21 and Amazon Titan models via bedrock api
awsyshah May 8, 2024
076ab2f
fix: response type for diffrent models and use of constant for top_P
awsyshah May 9, 2024
57a7653
fix: constant for top_P as int vs string
awsyshah May 10, 2024
a250277
feat: moved topP and maxTokens to config rather than being constants …
awsyshah Jul 12, 2024
e2359c7
Merge branch 'k8sgpt-ai:main' into Issue1076
awsyshah Jul 12, 2024
1e69355
Merge branch 'main' into Issue1076
AlexsJones Aug 13, 2024
2a40ab1
Merge branch 'main' into Issue1076
AlexsJones Sep 2, 2024
6148bf9
Merge branch 'main' into Issue1076
AlexsJones Sep 17, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
44 changes: 37 additions & 7 deletions pkg/ai/amazonbedrock.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,12 +52,18 @@ const (
ModelAnthropicClaudeV2 = "anthropic.claude-v2"
ModelAnthropicClaudeV1 = "anthropic.claude-v1"
ModelAnthropicClaudeInstantV1 = "anthropic.claude-instant-v1"
ModelA21J2UltraV1 = "ai21.j2-ultra-v1"
ModelA21J2JumboInstruct = "ai21.j2-jumbo-instruct"
ModelAmazonTitanExpressV1 = "amazon.titan-text-express-v1"
)

var BEDROCK_MODELS = []string{
ModelAnthropicClaudeV2,
ModelAnthropicClaudeV1,
ModelAnthropicClaudeInstantV1,
ModelA21J2UltraV1,
ModelA21J2JumboInstruct,
ModelAmazonTitanExpressV1,
}

// GetModelOrDefault check config model
Expand Down Expand Up @@ -116,13 +122,37 @@ func (a *AmazonBedRockClient) Configure(config IAIConfig) error {
// GetCompletion sends a request to the model for generating completion based on the provided prompt.
func (a *AmazonBedRockClient) GetCompletion(ctx context.Context, prompt string) (string, error) {

// Prepare the input data for the model invocation
request := map[string]interface{}{
"prompt": fmt.Sprintf("\n\nHuman: %s \n\nAssistant:", prompt),
"max_tokens_to_sample": 1024,
"temperature": a.temperature,
"top_p": 0.9,
}
// Prepare the input data for the model invocation based on the model
var request map[string]interface{}

switch a.model {
case ModelAnthropicClaudeV2, ModelAnthropicClaudeV1, ModelAnthropicClaudeInstantV1:
request = map[string]interface{}{
"prompt": fmt.Sprintf("\n\nHuman: %s \n\nAssistant:", prompt),
awsyshah marked this conversation as resolved.
Show resolved Hide resolved
"max_tokens_to_sample": 1024,
awsyshah marked this conversation as resolved.
Show resolved Hide resolved
"temperature": a.temperature,
"top_p": 0.9,
}
case ModelA21J2UltraV1, ModelA21J2JumboInstruct:
request = map[string]interface{}{
"prompt": fmt.Sprintf("\n\nHuman: %s \n\nAssistant:", prompt),
"maxTokens": 1024,
"temperature": a.temperature,
"topP": 0.9,
}
case ModelAmazonTitanExpressV1:
request = map[string]interface{}{
"inputText": fmt.Sprintf("\n\nHuman: %s \n\nAssistant:", prompt),
"textGenerationConfig": map[string]interface{}{
"maxTokenCount": 1024,
"temperature": a.temperature,
"topP": 0.9,
},
}
default:
return "", fmt.Errorf("model %s not supported", a.model)
}


body, err := json.Marshal(request)
if err != nil {
Expand Down
Loading