evan/aethera
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(chat): redesign input bar, expand sidebar, add thinking toggle

Browse Source 
- Restructure floating input: dominant textarea with compact bottom
  toolbar (model badge, thinking toggle, attach, send/stop).
- Model badge sizes to the current selection (not widest option) via
  a layered transparent select, with truncate-on-overflow fallback.
- Auto-expand the conversation sidebar on desktop and slide chat
  content right when open instead of overlaying.
- Add per-request thinking toggle (brain icon, default on, persisted
  in localStorage) sending chat_template_kwargs.enable_thinking.
- Always disable thinking for title summarization.
- Generate chat titles before the main response to keep the SSE
  stream from staying open past visible completion and to avoid
  busting the KV cache between turns.
This commit is contained in:
Evan Reichard
2026-05-17 16:36:11 -04:00
parent 8f732e6fc7
commit 6307a64c9c
7 changed files with 197 additions and 140 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
backend/internal/api/handlers.go
View File

@@ -345,7 +345,7 @@ func (a *API) PostChat(w http.ResponseWriter, r *http.Request) {
}
// Start Message
chunk, err := a.startMessageGeneration(chat.ID, genReq.Model, genReq.Prompt, genReq.Images)
chunk, err := a.startMessageGeneration(chat.ID, genReq.Model, genReq.Prompt, genReq.Images, genReq.EnableThinking())
if err != nil {
log.WithError(err).WithField("chat_id", chat.ID).Error("failed to start message generation")
http.Error(w, "Failed to start message generation", http.StatusInternalServerError)
@@ -539,7 +539,7 @@ func (a *API) PostChatMessage(w http.ResponseWriter, r *http.Request) {
}
// Start Message
chunk, err := a.startMessageGeneration(chatID, genReq.Model, genReq.Prompt, genReq.Images)
chunk, err := a.startMessageGeneration(chatID, genReq.Model, genReq.Prompt, genReq.Images, genReq.EnableThinking())
if err != nil {
log.WithError(err).WithField("chat_id", chatID).Error("failed to start message generation")
if errors.Is(err, errGenerationActive) {
@@ -572,7 +572,7 @@ func (a *API) getClient() (*client.Client, error) {
return a.client, nil
}
func (a *API) startMessageGeneration(chatID uuid.UUID, chatModel, userMessage string, images []string) (*MessageChunk, error) {
func (a *API) startMessageGeneration(chatID uuid.UUID, chatModel, userMessage string, images []string, enableThinking bool) (*MessageChunk, error) {
apiClient, err := a.getClient()
if err != nil {
return nil, fmt.Errorf("failed to get client: %w", err)
@@ -613,7 +613,7 @@ func (a *API) startMessageGeneration(chatID uuid.UUID, chatModel, userMessage st
}
return nil
}, func(gen *generation) {
a.runMessageGeneration(apiClient, chat, assistantMsg, chatModel, gen)
a.runMessageGeneration(apiClient, chat, assistantMsg, chatModel, enableThinking, gen)
}); err != nil {
return nil, err
}
@@ -621,13 +621,30 @@ func (a *API) startMessageGeneration(chatID uuid.UUID, chatModel, userMessage st
return initialChunk, nil
}
func (a *API) runMessageGeneration(apiClient *client.Client, chat *store.Chat, assistantMsg *store.Message, chatModel string, gen *generation) {
func (a *API) runMessageGeneration(apiClient *client.Client, chat *store.Chat, assistantMsg *store.Message, chatModel string, enableThinking bool, gen *generation) {
// Create Generation Context
ctx, cancel := context.WithTimeout(gen.ctx, a.textGenerationTimeout())
defer cancel()
// Generate Title First - Doing this before the main response avoids busting the KV
// cache between the user prompt and the assistant reply, and keeps the stream from
// staying open past the visible response completing.
if chat.Title == "" && len(chat.Messages) > 0 {
title, err := apiClient.CreateTitle(ctx, chat.Messages[0].Content, chatModel)
if err != nil {
a.logger.WithError(err).WithField("chat_id", chat.ID).Error("failed to create chat title")
} else {
chat.Title = title
if err := a.store.SaveChat(chat); err != nil {
a.logger.WithError(err).WithField("chat_id", chat.ID).Error("failed to update chat")
} else {
gen.broadcast(&MessageChunk{Chat: toChatNoMessages(chat)})
}
}
}
// Send Message
if _, err := apiClient.SendMessage(ctx, chat.Messages, chatModel, func(m *client.MessageChunk) error {
if _, err := apiClient.SendMessage(ctx, chat.Messages, chatModel, enableThinking, func(m *client.MessageChunk) error {
messageChanged := false
if m.Stats != nil {
@@ -684,17 +701,6 @@ func (a *API) runMessageGeneration(apiClient *client.Client, chat *store.Chat, a
return
}
// Summarize & Update Chat Title
if chat.Title == "" {
var err error
chat.Title, err = apiClient.CreateTitle(ctx, chat.Messages[0].Content, chatModel)
if err != nil {
a.logger.WithError(err).WithField("chat_id", chat.ID).Error("failed to create chat title")
} else if err := a.store.SaveChat(chat); err != nil {
a.logger.WithError(err).WithField("chat_id", chat.ID).Error("failed to update chat")
}
}
// Complete Assistant Message
assistantMsg.Status = store.MessageStatusComplete
if err := a.store.SaveChatMessage(assistantMsg); err != nil {

14
backend/internal/api/types.go
View File

@@ -69,9 +69,17 @@ type ImageRecord struct {
}
type GenerateTextRequest struct {
Model string `json:"model"`
Prompt string `json:"prompt"`
Images []string `json:"images,omitempty"`
Model string `json:"model"`
Prompt string `json:"prompt"`
Images []string `json:"images,omitempty"`
Thinking *bool `json:"thinking,omitempty"`
}
func (r *GenerateTextRequest) EnableThinking() bool {
if r.Thinking == nil {
return true
}
return *r.Thinking
}
func (r *GenerateTextRequest) Validate() error {

7
backend/internal/client/client.go
View File

@@ -66,7 +66,7 @@ func (c *Client) EditImage(ctx context.Context, body openai.ImageEditParams) ([]
return resp.Data, nil
}
func (c *Client) SendMessage(ctx context.Context, chatMessages []*store.Message, model string, cb StreamCallback) (string, error) {
func (c *Client) SendMessage(ctx context.Context, chatMessages []*store.Message, model string, enableThinking bool, cb StreamCallback) (string, error) {
// Ensure Callback
if cb == nil {
cb = func(mc *MessageChunk) error { return nil }
@@ -89,7 +89,8 @@ func (c *Client) SendMessage(ctx context.Context, chatMessages []*store.Message,
},
}
chatReq.SetExtraFields(map[string]any{
"timings_per_token": true, // Llama.cpp
"timings_per_token": true, // Llama.cpp
"chat_template_kwargs": map[string]any{"enable_thinking": enableThinking},
})
// Perform Request & Allocate Stats
@@ -172,7 +173,7 @@ func (c *Client) CreateTitle(ctx context.Context, userMessage, model string) (st
output, err := c.SendMessage(ctx, []*store.Message{{
Role: "user",
Content: prompt,
}}, model, nil)
}}, model, false, nil)
if err != nil {
return "", fmt.Errorf("failed to sent message: %w", err)
}

4
backend/internal/client/client_test.go
View File

@@ -32,7 +32,7 @@ func TestSendMessage(t *testing.T) {
_, err = client.SendMessage(ctx, []*store.Message{{
Role: "user",
Content: "What is 2+2? Think step by step.",
}}, model, func(mc *MessageChunk) error {
}}, model, true, func(mc *MessageChunk) error {
if mc.Thinking != nil {
_, err := thinkingBuf.Write([]byte(*mc.Thinking))
return err
@@ -118,7 +118,7 @@ func TestSendMessageWithImage(t *testing.T) {
Role: "user",
Content: "Describe this image in detail.",
Images: []string{dataURL},
}}, model, func(mc *MessageChunk) error {
}}, model, true, func(mc *MessageChunk) error {
if mc.Message != nil {
outputBuf.WriteString(*mc.Message)
}