steve
cbaf41f50c
Introduces an opt-in level-based reasoning toggle (low/medium/high) that each provider translates to its native parameter: - Anthropic: thinking.budget_tokens (1024/8000/24000), with temperature forced to default and MaxTokens auto-grown above the budget. - OpenAI/xAI/Groq via openaicompat: reasoning_effort string, gated by a new Rules.SupportsReasoning predicate so non-reasoning models don't receive the parameter. xAI uses Rules.MapReasoningEffort to remap "medium" to "high" since its API only accepts low|high. - Google: thinking_config.thinking_budget + include_thoughts:true. - DeepSeek: SupportsReasoning=false (reasoner is always-on; the reasoning_content trace was already extracted via openaicompat). Reasoning content is surfaced as Response.Thinking on Complete and as StreamEventThinking deltas during streaming. Provider-side: extracted from Anthropic thinking content blocks, Google's part.Thought=true parts, and the non-standard reasoning_content field that DeepSeek and Groq emit (parsed out of raw JSON since openai-go doesn't type it). Public API: - llm.ReasoningLevel + ReasoningLow/Medium/High constants - llm.WithReasoning(level) request option - Model.WithReasoning(level) for baked-in defaults - provider.Request.Reasoning, provider.Response.Thinking - provider.StreamEventThinking Tests cover Rules-based gating, MapReasoningEffort, reasoning_content extraction (Complete + Stream), Anthropic budget mapping, and temperature suppression when thinking is enabled. Existing behavior is unchanged when Reasoning is the empty string. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
39 lines
1.4 KiB
Go
39 lines
1.4 KiB
Go
// Package openai implements the go-llm v2 provider interface for OpenAI.
|
|
//
|
|
// The actual wire-protocol logic lives in the shared openaicompat package;
|
|
// this file encodes OpenAI-specific Rules (temperature is rejected on o-series
|
|
// and gpt-5* models) and supplies the default base URL.
|
|
package openai
|
|
|
|
import (
|
|
"strings"
|
|
|
|
"gitea.stevedudenhoeffer.com/steve/go-llm/v2/openaicompat"
|
|
)
|
|
|
|
// DefaultBaseURL is the public OpenAI Chat Completions endpoint.
|
|
const DefaultBaseURL = "https://api.openai.com/v1"
|
|
|
|
// Provider is the OpenAI chat-completion provider. It's a type alias over
|
|
// openaicompat.Provider so existing callers using openai.Provider keep compiling.
|
|
type Provider = openaicompat.Provider
|
|
|
|
// New creates a new OpenAI provider. An empty baseURL uses DefaultBaseURL.
|
|
func New(apiKey string, baseURL string) *Provider {
|
|
if baseURL == "" {
|
|
baseURL = DefaultBaseURL
|
|
}
|
|
return openaicompat.New(apiKey, baseURL, openaicompat.Rules{
|
|
RestrictTemperature: isReasoningModel,
|
|
SupportsReasoning: isReasoningModel,
|
|
})
|
|
}
|
|
|
|
// isReasoningModel reports whether the named OpenAI model is a reasoning
|
|
// model (o-series or gpt-5*). Reasoning models reject a user-supplied
|
|
// temperature and accept a reasoning_effort parameter; everything else
|
|
// rejects reasoning_effort.
|
|
func isReasoningModel(model string) bool {
|
|
return strings.HasPrefix(model, "o") || strings.HasPrefix(model, "gpt-5")
|
|
}
|