opencode/picoclaw
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix: only send prompt_cache_key for OpenAI provider

Browse source 
The prompt_cache_key field is an OpenAI-specific feature for prefix-based
prompt caching. When sent to other providers like Gemini that use the
OpenAI-compatible HTTP provider, it causes a 400 error because Gemini's
API rejects unknown fields.

This adds a supportPromptCache flag to the provider struct, enabled only
for the OpenAI protocol, so the field is no longer sent to providers
that don't support it.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Ubuntu 2026-02-26 03:00:10 +00:00
parent 094d65916d
commit bee6f75b2c
3 changed files with 24 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
pkg/providers/factory_provider.go
View file

@ -84,7 +84,9 @@ func CreateProviderFromConfig(cfg *config.ModelConfig) (LLMProvider, string, err
if apiBase == "" {
apiBase = getDefaultAPIBase(protocol)
}
return NewHTTPProviderWithMaxTokensField(cfg.APIKey, apiBase, cfg.Proxy, cfg.MaxTokensField), modelID, nil
p := NewHTTPProviderWithMaxTokensField(cfg.APIKey, apiBase, cfg.Proxy, cfg.MaxTokensField)
p.SetSupportPromptCache(true) // OpenAI supports prompt_cache_key
return p, modelID, nil
case "openrouter", "groq", "zhipu", "gemini", "nvidia",
"ollama", "moonshot", "shengsuanyun", "deepseek", "cerebras",

5
pkg/providers/http_provider.go
View file

@ -38,6 +38,11 @@ func (p *HTTPProvider) Chat(
return p.delegate.Chat(ctx, messages, tools, model, options)
}
// SetSupportPromptCache enables or disables sending the prompt_cache_key field.
func (p *HTTPProvider) SetSupportPromptCache(v bool) {
p.delegate.SetSupportPromptCache(v)
}
func (p *HTTPProvider) GetDefaultModel() string {
return ""
}

22
pkg/providers/openai_compat/provider.go
View file

@ -28,10 +28,11 @@ type (
)
type Provider struct {
apiKey string
apiBase string
maxTokensField string // Field name for max tokens (e.g., "max_completion_tokens" for o1/glm models)
httpClient *http.Client
apiKey string
apiBase string
maxTokensField string // Field name for max tokens (e.g., "max_completion_tokens" for o1/glm models)
httpClient *http.Client
supportPromptCache bool // Only send prompt_cache_key when true (OpenAI-specific feature)
}
func NewProvider(apiKey, apiBase, proxy string) *Provider {
@ -62,6 +63,12 @@ func NewProviderWithMaxTokensField(apiKey, apiBase, proxy, maxTokensField string
}
}
// SetSupportPromptCache enables or disables sending the prompt_cache_key field.
// Only OpenAI supports this field; other providers (e.g. Gemini) reject unknown fields.
func (p *Provider) SetSupportPromptCache(v bool) {
p.supportPromptCache = v
}
func (p *Provider) Chat(
ctx context.Context,
messages []Message,
@ -115,8 +122,11 @@ func (p *Provider) Chat(
// with the same key and reuse prefix KV cache across calls.
// The key is typically the agent ID — stable per agent, shared across requests.
// See: https://platform.openai.com/docs/guides/prompt-caching
if cacheKey, ok := options["prompt_cache_key"].(string); ok && cacheKey != "" {
requestBody["prompt_cache_key"] = cacheKey
// Only sent for providers that support it (e.g. OpenAI); others like Gemini reject unknown fields.
if p.supportPromptCache {
if cacheKey, ok := options["prompt_cache_key"].(string); ok && cacheKey != "" {
requestBody["prompt_cache_key"] = cacheKey
}
}
jsonData, err := json.Marshal(requestBody)