diff options
Diffstat (limited to 'models/models.go')
| -rw-r--r-- | models/models.go | 160 |
1 files changed, 156 insertions, 4 deletions
diff --git a/models/models.go b/models/models.go index 9514741..bb3fa1b 100644 --- a/models/models.go +++ b/models/models.go @@ -2,7 +2,6 @@ package models import ( "fmt" - "gf-lt/config" "strings" ) @@ -56,7 +55,7 @@ type RoleMsg struct { Content string `json:"content"` } -func (m RoleMsg) ToText(i int, cfg *config.Config) string { +func (m RoleMsg) ToText(i int) string { icon := fmt.Sprintf("(%d)", i) // check if already has role annotation (/completion makes them) if !strings.HasPrefix(m.Content, m.Role+":") { @@ -185,7 +184,7 @@ type DSCompletionReq struct { TopP float32 `json:"top_p"` } -func NewDSCompletionReq(prompt, model string, temp float32, cfg *config.Config, stopSlice []string) DSCompletionReq { +func NewDSCompletionReq(prompt, model string, temp float32, stopSlice []string) DSCompletionReq { return DSCompletionReq{ Model: model, Prompt: prompt, @@ -334,7 +333,7 @@ type LlamaCPPReq struct { // Samplers string `json:"samplers"` } -func NewLCPReq(prompt string, cfg *config.Config, props map[string]float32, stopStrings []string) LlamaCPPReq { +func NewLCPReq(prompt string, props map[string]float32, stopStrings []string) LlamaCPPReq { return LlamaCPPReq{ Stream: true, Prompt: prompt, @@ -362,3 +361,156 @@ type DSBalance struct { ToppedUpBalance string `json:"topped_up_balance"` } `json:"balance_infos"` } + +// openrouter +// https://openrouter.ai/docs/api-reference/completion +type OpenRouterCompletionReq struct { + Model string `json:"model"` + Prompt string `json:"prompt"` + Stream bool `json:"stream"` + Temperature float32 `json:"temperature"` + Stop []string `json:"stop"` // not present in docs + MinP float32 `json:"min_p"` + NPredict int32 `json:"max_tokens"` +} + +func NewOpenRouterCompletionReq(model, prompt string, props map[string]float32, stopStrings []string) OpenRouterCompletionReq { + return OpenRouterCompletionReq{ + Stream: true, + Prompt: prompt, + Temperature: props["temperature"], + MinP: props["min_p"], + NPredict: int32(props["n_predict"]), + Stop: stopStrings, + Model: model, + } +} + +type OpenRouterChatReq struct { + Messages []RoleMsg `json:"messages"` + Model string `json:"model"` + Stream bool `json:"stream"` + Temperature float32 `json:"temperature"` + MinP float32 `json:"min_p"` + NPredict int32 `json:"max_tokens"` +} + +func NewOpenRouterChatReq(cb ChatBody, props map[string]float32) OpenRouterChatReq { + return OpenRouterChatReq{ + Messages: cb.Messages, + Model: cb.Model, + Stream: cb.Stream, + Temperature: props["temperature"], + MinP: props["min_p"], + NPredict: int32(props["n_predict"]), + } +} + +type OpenRouterChatRespNonStream struct { + ID string `json:"id"` + Provider string `json:"provider"` + Model string `json:"model"` + Object string `json:"object"` + Created int `json:"created"` + Choices []struct { + Logprobs any `json:"logprobs"` + FinishReason string `json:"finish_reason"` + NativeFinishReason string `json:"native_finish_reason"` + Index int `json:"index"` + Message struct { + Role string `json:"role"` + Content string `json:"content"` + Refusal any `json:"refusal"` + Reasoning any `json:"reasoning"` + } `json:"message"` + } `json:"choices"` + Usage struct { + PromptTokens int `json:"prompt_tokens"` + CompletionTokens int `json:"completion_tokens"` + TotalTokens int `json:"total_tokens"` + } `json:"usage"` +} + +type OpenRouterChatResp struct { + ID string `json:"id"` + Provider string `json:"provider"` + Model string `json:"model"` + Object string `json:"object"` + Created int `json:"created"` + Choices []struct { + Index int `json:"index"` + Delta struct { + Role string `json:"role"` + Content string `json:"content"` + } `json:"delta"` + FinishReason string `json:"finish_reason"` + NativeFinishReason string `json:"native_finish_reason"` + Logprobs any `json:"logprobs"` + } `json:"choices"` +} + +type OpenRouterCompletionResp struct { + ID string `json:"id"` + Provider string `json:"provider"` + Model string `json:"model"` + Object string `json:"object"` + Created int `json:"created"` + Choices []struct { + Text string `json:"text"` + FinishReason string `json:"finish_reason"` + NativeFinishReason string `json:"native_finish_reason"` + Logprobs any `json:"logprobs"` + } `json:"choices"` +} + +type ORModel struct { + ID string `json:"id"` + CanonicalSlug string `json:"canonical_slug"` + HuggingFaceID string `json:"hugging_face_id"` + Name string `json:"name"` + Created int `json:"created"` + Description string `json:"description"` + ContextLength int `json:"context_length"` + Architecture struct { + Modality string `json:"modality"` + InputModalities []string `json:"input_modalities"` + OutputModalities []string `json:"output_modalities"` + Tokenizer string `json:"tokenizer"` + InstructType any `json:"instruct_type"` + } `json:"architecture"` + Pricing struct { + Prompt string `json:"prompt"` + Completion string `json:"completion"` + Request string `json:"request"` + Image string `json:"image"` + Audio string `json:"audio"` + WebSearch string `json:"web_search"` + InternalReasoning string `json:"internal_reasoning"` + } `json:"pricing,omitempty"` + TopProvider struct { + ContextLength int `json:"context_length"` + MaxCompletionTokens int `json:"max_completion_tokens"` + IsModerated bool `json:"is_moderated"` + } `json:"top_provider"` + PerRequestLimits any `json:"per_request_limits"` + SupportedParameters []string `json:"supported_parameters"` +} + +type ORModels struct { + Data []ORModel `json:"data"` +} + +func (orm *ORModels) ListModels(free bool) []string { + resp := []string{} + for _, model := range orm.Data { + if free { + if model.Pricing.Prompt == "0" && model.Pricing.Request == "0" && + model.Pricing.Completion == "0" { + resp = append(resp, model.ID) + } + } else { + resp = append(resp, model.ID) + } + } + return resp +} |