inference

package

v0.1.0 Latest Latest Go to latest Published: Aug 9, 2026 License: Apache-2.0 Imports: 16 Imported by: 0

Details

Valid go.mod file
Redistributable license
Tagged version
Stable version
Learn more about best practices

Repository

github.com/ryanaldo34/tacklr

Links

Open Source Insights

Documentation ¶

Index ¶

type APIStatusError
type OpenAIInferenceStrategy
- func NewOpenAIInferenceStrategy(client *http.Client) *OpenAIInferenceStrategy

Constants ¶

This section is empty.

Variables ¶

This section is empty.

Functions ¶

This section is empty.

Types ¶

type APIStatusError ¶

type APIStatusError struct {
	Status int
	Body   string
	Code   string
}

APIStatusError conveys an HTTP error response from an upstream LLM API. Use errors.As to extract structured status/body details from a wrapped chain.

func (*APIStatusError) Error ¶

func (e *APIStatusError) Error() string

func (*APIStatusError) ProviderErrorCode ¶

func (e *APIStatusError) ProviderErrorCode() string

ProviderErrorCode implements tacklr.ProviderStatus.

func (*APIStatusError) ProviderHTTPStatus ¶

func (e *APIStatusError) ProviderHTTPStatus() int

ProviderHTTPStatus implements tacklr.ProviderStatus.

type OpenAIInferenceStrategy ¶

type OpenAIInferenceStrategy struct {
	// contains filtered or unexported fields
}

func NewOpenAIInferenceStrategy ¶

func NewOpenAIInferenceStrategy(client *http.Client) *OpenAIInferenceStrategy

func (*OpenAIInferenceStrategy) CompressContextWindow ¶

func (s *OpenAIInferenceStrategy) CompressContextWindow() error

func (*OpenAIInferenceStrategy) CountTokens ¶

func (s *OpenAIInferenceStrategy) CountTokens(ctx context.Context, messages []*tacklr.Message, tools []*tacklr.Tool) (int, error)

func (*OpenAIInferenceStrategy) Invoke ¶

func (s *OpenAIInferenceStrategy) Invoke(ctx context.Context, messages []*tacklr.Message, tools []*tacklr.Tool) (chan tacklr.LLMResponseChunk, error)

func (*OpenAIInferenceStrategy) MaxContextWindow ¶

func (s *OpenAIInferenceStrategy) MaxContextWindow() (int, error)

func (*OpenAIInferenceStrategy) ModelTelemetryIdentity ¶

func (s *OpenAIInferenceStrategy) ModelTelemetryIdentity() telemetry.ModelIdentity

ModelTelemetryIdentity implements the optional harness hook so model spans get GenAI provider/model attrs without exporting raw config fields.

func (*OpenAIInferenceStrategy) SetSystemPrompt ¶

func (s *OpenAIInferenceStrategy) SetSystemPrompt(prompt string)

func (*OpenAIInferenceStrategy) SupportsMIME ¶

func (s *OpenAIInferenceStrategy) SupportsMIME(mimeType string) bool

SupportsMIME implements tacklr.InferenceStrategy for the selected model id.

func (*OpenAIInferenceStrategy) WithApiKey ¶

func (s *OpenAIInferenceStrategy) WithApiKey(key string) tacklr.InferenceStrategy

func (*OpenAIInferenceStrategy) WithMaxOutputTokens ¶

func (s *OpenAIInferenceStrategy) WithMaxOutputTokens(n int) *OpenAIInferenceStrategy

WithMaxOutputTokens sets Responses API max_output_tokens (0 omits the field). Raise this for Azure reasoning models after large tool results (e.g. web_search) so streams do not end as bare response.incomplete.

func (*OpenAIInferenceStrategy) WithModel ¶

func (s *OpenAIInferenceStrategy) WithModel(model string) tacklr.InferenceStrategy

func (*OpenAIInferenceStrategy) WithReasoningLevel ¶

func (s *OpenAIInferenceStrategy) WithReasoningLevel(level string) tacklr.InferenceStrategy

func (*OpenAIInferenceStrategy) WithReasoningSummary ¶

func (s *OpenAIInferenceStrategy) WithReasoningSummary(summary string) *OpenAIInferenceStrategy

WithReasoningSummary sets reasoning.summary on Responses API requests ("auto", "concise", "detailed"). Empty clears it.

func (*OpenAIInferenceStrategy) WithStructuredOutput ¶

func (s *OpenAIInferenceStrategy) WithStructuredOutput(v any) tacklr.InferenceStrategy

func (*OpenAIInferenceStrategy) WithURL ¶

func (s *OpenAIInferenceStrategy) WithURL(url string) tacklr.InferenceStrategy

Source Files ¶

View all Source files

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL