From 2ff83724395679b51030fed56cdda185aec3683a Mon Sep 17 00:00:00 2001 From: Yiyang Kang Date: Sun, 2 Feb 2025 15:26:07 +0900 Subject: [PATCH 1/3] build: update dependencies --- go.mod | 12 ++++++------ go.sum | 12 ++++++++++++ 2 files changed, 18 insertions(+), 6 deletions(-) diff --git a/go.mod b/go.mod index 324a7f5..84d94c4 100644 --- a/go.mod +++ b/go.mod @@ -11,10 +11,10 @@ require ( github.com/eko/gocache/lib/v4 v4.2.0 github.com/eko/gocache/store/ristretto/v4 v4.2.2 github.com/go-errors/errors v1.5.1 - github.com/go-resty/resty/v2 v2.16.3 - github.com/goccy/go-json v0.10.4 + github.com/go-resty/resty/v2 v2.16.5 + github.com/goccy/go-json v0.10.5 github.com/ilyakaznacheev/cleanenv v1.5.0 - github.com/samber/lo v1.47.0 + github.com/samber/lo v1.49.1 go.uber.org/zap v1.27.0 golang.org/x/net v0.34.0 gopkg.in/telebot.v3 v3.3.8 @@ -32,15 +32,15 @@ require ( github.com/pkg/errors v0.9.1 // indirect github.com/prometheus/client_golang v1.20.5 // indirect github.com/prometheus/client_model v0.6.1 // indirect - github.com/prometheus/common v0.61.0 // indirect + github.com/prometheus/common v0.62.0 // indirect github.com/prometheus/procfs v0.15.1 // indirect go.uber.org/mock v0.5.0 // indirect go.uber.org/multierr v1.11.0 // indirect - golang.org/x/exp v0.0.0-20250106191152-7588d65b2ba8 // indirect + golang.org/x/exp v0.0.0-20250128182459-e0ece0dbea4c // indirect golang.org/x/sync v0.10.0 // indirect golang.org/x/sys v0.29.0 // indirect golang.org/x/text v0.21.0 // indirect - google.golang.org/protobuf v1.36.2 // indirect + google.golang.org/protobuf v1.36.4 // indirect gopkg.in/yaml.v3 v3.0.1 // indirect olympos.io/encoding/edn v0.0.0-20201019073823-d3554ca0b0a3 // indirect ) diff --git a/go.sum b/go.sum index d4b7c85..a5270b2 100644 --- a/go.sum +++ b/go.sum @@ -157,9 +157,13 @@ github.com/go-playground/universal-translator v0.17.0/go.mod h1:UkSxE5sNxxRwHyU+ github.com/go-playground/validator/v10 v10.4.1/go.mod h1:nlOn6nFhuKACm19sB/8EGNn9GlaMV7XkbRSipzJ0Ii4= github.com/go-resty/resty/v2 v2.16.3 h1:zacNT7lt4b8M/io2Ahj6yPypL7bqx9n1iprfQuodV+E= github.com/go-resty/resty/v2 v2.16.3/go.mod h1:hkJtXbA2iKHzJheXYvQ8snQES5ZLGKMwQ07xAwp/fiA= +github.com/go-resty/resty/v2 v2.16.5 h1:hBKqmWrr7uRc3euHVqmh1HTHcKn99Smr7o5spptdhTM= +github.com/go-resty/resty/v2 v2.16.5/go.mod h1:hkJtXbA2iKHzJheXYvQ8snQES5ZLGKMwQ07xAwp/fiA= github.com/go-stack/stack v1.8.0/go.mod h1:v0f6uXyyMGvRgIKkXu+yp6POWl0qKG85gN/melR3HDY= github.com/goccy/go-json v0.10.4 h1:JSwxQzIqKfmFX1swYPpUThQZp/Ka4wzJdK0LWVytLPM= github.com/goccy/go-json v0.10.4/go.mod h1:oq7eo15ShAhp70Anwd5lgX2pLfOS3QCiwU/PULtXL6M= +github.com/goccy/go-json v0.10.5 h1:Fq85nIqj+gXn/S5ahsiTlK3TmC85qgirsdTP/+DeaC4= +github.com/goccy/go-json v0.10.5/go.mod h1:oq7eo15ShAhp70Anwd5lgX2pLfOS3QCiwU/PULtXL6M= github.com/goccy/go-yaml v1.9.5/go.mod h1:U/jl18uSupI5rdI2jmuCswEA2htH9eXfferR3KfscvA= github.com/godbus/dbus/v5 v5.0.4/go.mod h1:xhWf0FNVPg57R7Z0UbKHbJfkEywrmjJnf7w5xrFpKfA= github.com/gogo/protobuf v1.1.1/go.mod h1:r8qH/GZQm5c6nD/R0oafs1akxWv10x8SbQlK7atdtwQ= @@ -375,6 +379,8 @@ github.com/prometheus/common v0.10.0/go.mod h1:Tlit/dnDKsSWFlCLTWaA1cyBgKHSMdTB8 github.com/prometheus/common v0.26.0/go.mod h1:M7rCNAaPfAosfx8veZJCuw84e35h3Cfd9VFqTh1DIvc= github.com/prometheus/common v0.61.0 h1:3gv/GThfX0cV2lpO7gkTUwZru38mxevy90Bj8YFSRQQ= github.com/prometheus/common v0.61.0/go.mod h1:zr29OCN/2BsJRaFwG8QOBr41D6kkchKbpeNH7pAjb/s= +github.com/prometheus/common v0.62.0 h1:xasJaQlnWAeyHdUBeGjXmutelfJHWMRr+Fg4QszZ2Io= +github.com/prometheus/common v0.62.0/go.mod h1:vyBcEuLSvWos9B1+CyL7JZ2up+uFzXhkqml0W5zIY1I= github.com/prometheus/procfs v0.0.0-20181005140218-185b4288413d/go.mod h1:c3At6R/oaqEKCNdg8wHV1ftS6bRYblBhIjjI8uT2IGk= github.com/prometheus/procfs v0.0.2/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsTZCD3I8kEA= github.com/prometheus/procfs v0.0.8/go.mod h1:7Qr8sr6344vo1JqZ6HhLceV9o3AJ1Ff+GxbHq6oeK9A= @@ -391,6 +397,8 @@ github.com/ryanuber/columnize v0.0.0-20160712163229-9b3edd62028f/go.mod h1:sm1tb github.com/sagikazarmark/crypt v0.6.0/go.mod h1:U8+INwJo3nBv1m6A/8OBXAq7Jnpspk5AxSgDyEQcea8= github.com/samber/lo v1.47.0 h1:z7RynLwP5nbyRscyvcD043DWYoOcYRv3mV8lBeqOCLc= github.com/samber/lo v1.47.0/go.mod h1:RmDH9Ct32Qy3gduHQuKJ3gW1fMHAnE/fAzQuf6He5cU= +github.com/samber/lo v1.49.1 h1:4BIFyVfuQSEpluc7Fua+j1NolZHiEHEpaSEKdsH0tew= +github.com/samber/lo v1.49.1/go.mod h1:dO6KHFzUKXgP8LDhU0oI8d2hekjXnGOu0DB8Jecxd6o= github.com/sean-/seed v0.0.0-20170313163322-e2103e2c3529/go.mod h1:DxrIzT+xaE7yg65j358z/aeFdxmN0P9QXhEzd20vsDc= github.com/sirupsen/logrus v1.2.0/go.mod h1:LxeOpSwHxABJmUn/MG1IvRgCAasNZTLOkJPxbbu5VWo= github.com/sirupsen/logrus v1.4.2/go.mod h1:tLMulIdttU9McNUspp0xgXVQah82FyeX6MwdIuYE2rE= @@ -473,6 +481,8 @@ golang.org/x/exp v0.0.0-20200207192155-f17229e696bd/go.mod h1:J/WKrq2StrnmMY6+EH golang.org/x/exp v0.0.0-20200224162631-6cc2880d07d6/go.mod h1:3jZMyOhIsHpP37uCMkUooju7aAi5cS1Q23tOzKc+0MU= golang.org/x/exp v0.0.0-20250106191152-7588d65b2ba8 h1:yqrTHse8TCMW1M1ZCP+VAR/l0kKxwaAIqN/il7x4voA= golang.org/x/exp v0.0.0-20250106191152-7588d65b2ba8/go.mod h1:tujkw807nyEEAamNbDrEGzRav+ilXA7PCRAd6xsmwiU= +golang.org/x/exp v0.0.0-20250128182459-e0ece0dbea4c h1:KL/ZBHXgKGVmuZBZ01Lt57yE5ws8ZPSkkihmEyq7FXc= +golang.org/x/exp v0.0.0-20250128182459-e0ece0dbea4c/go.mod h1:tujkw807nyEEAamNbDrEGzRav+ilXA7PCRAd6xsmwiU= golang.org/x/image v0.0.0-20190227222117-0694c2d4d067/go.mod h1:kZ7UVZpmo3dzQBMxlp+ypCbDeSB+sBbTgSJuh5dn5js= golang.org/x/image v0.0.0-20190802002840-cff245a6509b/go.mod h1:FeLwcggjj3mMvU+oOTbSwawSJRM1uh48EjtB4UJZlP0= golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE= @@ -951,6 +961,8 @@ google.golang.org/protobuf v1.27.1/go.mod h1:9q0QmTI4eRPtz6boOQmLYwt+qCgq0jsYwAQ google.golang.org/protobuf v1.28.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I= google.golang.org/protobuf v1.36.2 h1:R8FeyR1/eLmkutZOM5CWghmo5itiG9z0ktFlTVLuTmU= google.golang.org/protobuf v1.36.2/go.mod h1:9fA7Ob0pmnwhb644+1+CVWFRbNajQ6iRojtC/QF5bRE= +google.golang.org/protobuf v1.36.4 h1:6A3ZDJHn/eNqc1i+IdefRzy/9PokBTPvcqMySR7NNIM= +google.golang.org/protobuf v1.36.4/go.mod h1:9fA7Ob0pmnwhb644+1+CVWFRbNajQ6iRojtC/QF5bRE= gopkg.in/alecthomas/kingpin.v2 v2.2.6/go.mod h1:FMv+mEhP44yOT+4EoQTLFTRgOQ1FBLkstjWtayDeSgw= gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0= gopkg.in/check.v1 v1.0.0-20180628173108-788fd7840127/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0= From 4ab359053a6811e9ce9a7db2aba32aa9d0456e27 Mon Sep 17 00:00:00 2001 From: Yiyang Kang Date: Sun, 2 Feb 2025 15:28:16 +0900 Subject: [PATCH 2/3] feat: update according to openai doc --- assistant.go | 5 ++--- openai/chat.go | 36 ++++++++++++++++++++++++------------ openai/client.go | 2 +- openai/models.go | 5 +++-- 4 files changed, 30 insertions(+), 18 deletions(-) diff --git a/assistant.go b/assistant.go index 33e7bd0..081e5bc 100644 --- a/assistant.go +++ b/assistant.go @@ -193,8 +193,8 @@ func handleAssistantConversation(c tele.Context, thread []*tele.Message) error { logger.Warnw("failed to cache message", "error", err) } - nBytes := 0 // Used to estimated number of tokens. For now we treat 3 bytes as 1 token. - nBytesMax := (4096 - 512) * 3 // Leave some space for the response + nBytes := 0 // Used to estimated number of tokens. For now we treat 3 bytes as 1 token. + nBytesMax := 16384 * 3 // Leave some space for the response sysMsg := prompts.Assistant() chatReqMsgs := []openai.ChatMessage{ @@ -237,7 +237,6 @@ func handleAssistantConversation(c tele.Context, thread []*tele.Message) error { Model: openai.ModelGpt4O, Messages: chatReqMsgs, Temperature: lo.ToPtr(0.42), - MaxTokens: 2048, User: assistantHashUserId(lastMsg.Sender.ID), } diff --git a/openai/chat.go b/openai/chat.go index 3564ed6..605f43c 100644 --- a/openai/chat.go +++ b/openai/chat.go @@ -6,28 +6,40 @@ type ChatRole string const ( ChatRoleSystem ChatRole = "system" + ChatRoleDeveloper ChatRole = "developer" // replaces `system` role for o1 and newer models + ChatRoleTool ChatRole = "tool" ChatRoleAssistant ChatRole = "assistant" ChatRoleUser ChatRole = "user" ) +type ReasoningEffort string + +const ( + ReasoningEffortLow ReasoningEffort = "low" + ReasoningEffortMedium ReasoningEffort = "medium" + ReasoningEffortHigh ReasoningEffort = "high" +) + type ChatMessage struct { Role ChatRole `json:"role"` Content string `json:"content"` } type ChatRequest struct { - Model string `json:"model"` - Messages []ChatMessage `json:"messages"` - Temperature *float64 `json:"temperature,omitempty"` // What sampling temperature to use, between 0 and 2. - TopP *float64 `json:"top_p,omitempty"` // Nucleus sampling. Specify this or temperature but not both. - N int `json:"n,omitempty"` // How many chat completion choices to generate for each input message. - Stream bool `json:"stream,omitempty"` // If set, partial message deltas will be sent as data-only server-sent events as they become available. - Stop []string `json:"stop,omitempty"` // Up to 4 sequences where the API will stop generating further tokens. - MaxTokens int `json:"max_tokens,omitempty"` - PresencePenalty *float64 `json:"presence_penalty,omitempty"` // Number between -2.0 and 2.0. - FrequencyPenalty *float64 `json:"frequency_penalty,omitempty"` // Number between -2.0 and 2.0. - LogitBias map[string]float64 `json:"logit_bias,omitempty"` // Modify the likelihood of specified tokens appearing in the completion. - User string `json:"user,omitempty"` // A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. + Model string `json:"model"` + Messages []ChatMessage `json:"messages"` + Temperature *float64 `json:"temperature,omitempty"` // What sampling temperature to use, between 0 and 2. + TopP *float64 `json:"top_p,omitempty"` // Nucleus sampling. Specify this or temperature but not both. + N int `json:"n,omitempty"` // How many chat completion choices to generate for each input message. + Stream bool `json:"stream,omitempty"` // If set, partial message deltas will be sent as data-only server-sent events as they become available. + Stop []string `json:"stop,omitempty"` // Up to 4 sequences where the API will stop generating further tokens. + MaxTokens int `json:"max_tokens,omitempty"` // Deprecated: in favor of `max_completion_tokens` + MaxCompletionTokens int `json:"max_completion_tokens,omitempty"` // Including visible output tokens and reasoning tokens. + PresencePenalty *float64 `json:"presence_penalty,omitempty"` // Number between -2.0 and 2.0. + FrequencyPenalty *float64 `json:"frequency_penalty,omitempty"` // Number between -2.0 and 2.0. + LogitBias map[string]float64 `json:"logit_bias,omitempty"` // Modify the likelihood of specified tokens appearing in the completion. + User string `json:"user,omitempty"` // A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. + ReasoningEffort ReasoningEffort `json:"reasoning_effort,omitempty"` // Constrains effort on reasoning for reasoning models. } type ChatResponseChoice struct { diff --git a/openai/client.go b/openai/client.go index 1800cf9..2802647 100644 --- a/openai/client.go +++ b/openai/client.go @@ -20,7 +20,7 @@ func NewClient(apiKey string) *Client { cli := resty.New(). SetTransport(&http.Transport{ Proxy: http.ProxyFromEnvironment, - ResponseHeaderTimeout: 10 * time.Second, + ResponseHeaderTimeout: 90 * time.Second, }). SetBaseURL("https://api.openai.com"). SetHeader("Authorization", "Bearer "+apiKey). diff --git a/openai/models.go b/openai/models.go index 7398247..0307fcc 100644 --- a/openai/models.go +++ b/openai/models.go @@ -2,6 +2,7 @@ package openai const ( ModelGpt4O = "gpt-4o" // Safe default - ModelO1Preview = "o1-preview" // Expensive - ModelO1Mini = "o1-mini" + ModelO1Preview = "o1-preview" // Expensive reasoning model + ModelO1Mini = "o1-mini" // Cheaper reasoning model + ModelO3Mini = "o3-mini" // Cheaper yet powerful reasoning model ) From 6328ff0f188a33bde9fd4b167a426a77e94bf604 Mon Sep 17 00:00:00 2001 From: Yiyang Kang Date: Sun, 2 Feb 2025 15:30:08 +0900 Subject: [PATCH 3/3] feat: add reasoning command --- bot.go | 1 + botcmd_reason.go | 65 ++++++++++++++++++++++++++++++++++++++++++++++++ main.go | 1 + 3 files changed, 67 insertions(+) create mode 100644 botcmd_reason.go diff --git a/bot.go b/bot.go index b59b099..2818f4f 100644 --- a/bot.go +++ b/bot.go @@ -38,6 +38,7 @@ func initBot() (*tele.Bot, error) { b.Handle("/year_progress", handleYearProgressCmd) b.Handle("/xr", handleExchangeRateCmd) + b.Handle("/reason", handleReasonCmd) b.Handle("/tr", handleTranslateCmd) for _, tbtn := range translateBtns { b.Handle(tbtn, handleTranslateBtn) diff --git a/botcmd_reason.go b/botcmd_reason.go new file mode 100644 index 0000000..b0763e4 --- /dev/null +++ b/botcmd_reason.go @@ -0,0 +1,65 @@ +package main + +import ( + "regexp" + "strings" + + tele "gopkg.in/telebot.v3" + + "git.gensokyo.cafe/kkyy/tgbot_misaka_5882f7/openai" +) + +var ( + reasonCmdRe = regexp.MustCompile(`^\s*\/reason(@\S*)?\s*`) + reasoningIndicatorMessage = "🤔💭 Thinking..." +) + +func handleReasonCmd(c tele.Context) error { + msg := c.Message() + if msg == nil { + return nil + } + + payload := strings.TrimSpace(reasonCmdRe.ReplaceAllString(msg.Text, "")) + if payload == "" { + return c.Reply("Usage: `/reason `", + &tele.SendOptions{ParseMode: tele.ModeMarkdown}, + tele.Silent, + ) + } + + req := openai.ChatRequest{ + Model: openai.ModelO1Mini, + Messages: []openai.ChatMessage{ + { + Role: openai.ChatRoleUser, + Content: payload, + }, + }, + // reasoning_effort is only available to `o1` and `o3-mini`, which is not yet accessible. + // ReasoningEffort: openai.ReasoningEffortHigh, + } + + replyMsg, err := c.Bot().Reply(msg, reasoningIndicatorMessage, tele.Silent) + if err != nil { + logger.Errorw("assistant: failed to complete reasoning request", "error", err) + return c.Reply("Sorry, there's a technical issue. 😵💫 Please try again later.", tele.Silent) + } + err = assistantStreamedResponse(req, func(text string, finished bool) error { + var err error + replyMsg, err = c.Bot().Edit(replyMsg, text) + if finished && err == nil { + replyMsg.ReplyTo = msg // nasty bug + if err := cacheMessage(replyMsg); err != nil { + logger.Warnw("failed to cache message", "error", err) + } + } + return err + }) + + if err != nil { + logger.Errorw("assistant: failed to complete reasoning request", "error", err) + return c.Reply("Sorry, there's a technical issue. 😵💫 Please try again later.", tele.Silent) + } + return nil +} diff --git a/main.go b/main.go index b009728..38936ba 100644 --- a/main.go +++ b/main.go @@ -41,6 +41,7 @@ func runBot() { logger.Info("Announcing commands...") if err = bot.SetCommands([]tele.Command{ + {Text: "reason", Description: "Think."}, {Text: "tr", Description: "Translate text"}, {Text: "kanji", Description: "Help with pronunciation of Kanji"}, {Text: "xr", Description: "Currency exchange rates"},