feat: gpt-5.2, refactored streaming, improved md rendering, basic tools

internal/api/chat/list_supported_models_v2.go

@@ -10,6 +10,202 @@ import (
 	"github.com/openai/openai-go/v3"
 )
 
+// modelConfig holds model configuration including whether it requires user's own API key
+type modelConfig struct {
+	name           string
+	slugOpenRouter string // Slug for OpenRouter API (used when user doesn't provide own key)
+	slugOpenAI     string // Slug for OpenAI API (used when user provides own key)
+	totalContext   int64
+	maxOutput      int64
+	inputPrice     int64
+	outputPrice    int64
+	requireOwnKey  bool // If true, this model requires user to provide their own API key
+}
+
+// allModels defines all available models in the system
+var allModels = []modelConfig{
+	{
+		name:           "GPT-5.1",
+		slugOpenRouter: "openai/gpt-5.1",
+		slugOpenAI:     openai.ChatModelGPT5_1,
+		totalContext:   400000,
+		maxOutput:      128000,
+		inputPrice:     125,  // $1.25
+		outputPrice:    1000, // $10.00
+		requireOwnKey:  false,
+	},
+	{
+		name:           "GPT-5.2",
+		slugOpenRouter: "openai/gpt-5.2",
+		slugOpenAI:     openai.ChatModelGPT5_2,
+		totalContext:   400000,
+		maxOutput:      128000,
+		inputPrice:     175,  // $1.75
+		outputPrice:    1400, // $14.00
+		requireOwnKey:  true,
+	},
+	{
+		name:           "GPT-5 Mini",
+		slugOpenRouter: "openai/gpt-5-mini",
+		slugOpenAI:     openai.ChatModelGPT5Mini,
+		totalContext:   400000,
+		maxOutput:      128000,
+		inputPrice:     25,
+		outputPrice:    200,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "GPT-5 Nano",
+		slugOpenRouter: "openai/gpt-5-nano",
+		slugOpenAI:     openai.ChatModelGPT5Nano,
+		totalContext:   400000,
+		maxOutput:      128000,
+		inputPrice:     5,  // $0.20
+		outputPrice:    40, // $0.80
+		requireOwnKey:  false,
+	},
+	{
+		name:           "GPT-4.1",
+		slugOpenRouter: "openai/gpt-4.1",
+		slugOpenAI:     openai.ChatModelGPT4_1,
+		totalContext:   1050000,
+		maxOutput:      32800,
+		inputPrice:     200, // $2.00
+		outputPrice:    800,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "GPT-4.1-mini",
+		slugOpenRouter: "openai/gpt-4.1-mini",
+		slugOpenAI:     openai.ChatModelGPT4_1Mini,
+		totalContext:   128000,
+		maxOutput:      16400,
+		inputPrice:     15,
+		outputPrice:    60,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "GPT-4o",
+		slugOpenRouter: "openai/gpt-4o",
+		slugOpenAI:     openai.ChatModelGPT4o,
+		totalContext:   128000,
+		maxOutput:      16400,
+		inputPrice:     250,
+		outputPrice:    1000,
+		requireOwnKey:  true,
+	},
+	{
+		name:           "OpenAI: gpt-oss-120b (free)",
+		slugOpenRouter: "openai/gpt-oss-120b:free",
+		slugOpenAI:     "",
+		totalContext:   131072,
+		maxOutput:      131072,
+		inputPrice:     0,
+		outputPrice:    0,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "Qwen Plus (balanced)",
+		slugOpenRouter: "qwen/qwen-plus",
+		slugOpenAI:     "", // OpenAI doesn't support Qwen, use OpenRouter slug
+		totalContext:   131100,
+		maxOutput:      8200,
+		inputPrice:     40,
+		outputPrice:    120,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "Qwen Turbo (fast)",
+		slugOpenRouter: "qwen/qwen-turbo",
+		slugOpenAI:     "", // OpenAI doesn't support Qwen, use OpenRouter slug
+		totalContext:   1000000,
+		maxOutput:      8200,
+		inputPrice:     5,
+		outputPrice:    20,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "Qwen3 Coder 480B A35B (free)",
+		slugOpenRouter: "qwen/qwen3-coder:free",
+		slugOpenAI:     "",
+		totalContext:   262000,
+		maxOutput:      262000,
+		inputPrice:     0,
+		outputPrice:    0,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "GLM 4.5 Air (free)",
+		slugOpenRouter: "z-ai/glm-4.5-air:free",
+		slugOpenAI:     "",
+		totalContext:   131072,
+		maxOutput:      131072,
+		inputPrice:     0,
+		outputPrice:    0,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "Gemini 2.5 Flash (fast)",
+		slugOpenRouter: "google/gemini-2.5-flash",
+		slugOpenAI:     "", // OpenAI doesn't support Gemini, use OpenRouter slug
+		totalContext:   1050000,
+		maxOutput:      65500,
+		inputPrice:     30,
+		outputPrice:    250,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "Gemini 3 Flash Preview",
+		slugOpenRouter: "google/gemini-3-flash-preview",
+		slugOpenAI:     "", // OpenAI doesn't support Gemini, use OpenRouter slug
+		totalContext:   1050000,
+		maxOutput:      65500,
+		inputPrice:     50,
+		outputPrice:    300,
+		requireOwnKey:  false,
+	},
+	{
+		name:           "o1 Mini",
+		slugOpenRouter: "openai/o1-mini",
+		slugOpenAI:     openai.ChatModelO1Mini,
+		totalContext:   128000,
+		maxOutput:      65536,
+		inputPrice:     300,  // $3.00
+		outputPrice:    1200, // $12.00
+		requireOwnKey:  true,
+	},
+	{
+		name:           "o3",
+		slugOpenRouter: "openai/o3",
+		slugOpenAI:     openai.ChatModelO3,
+		totalContext:   200000,
+		maxOutput:      100000,
+		inputPrice:     200,
+		outputPrice:    800,
+		requireOwnKey:  true,
+	},
+	{
+		name:           "o3 Mini",
+		slugOpenRouter: "openai/o3-mini",
+		slugOpenAI:     openai.ChatModelO3Mini,
+		totalContext:   200000,
+		maxOutput:      100000,
+		inputPrice:     110,
+		outputPrice:    440,
+		requireOwnKey:  true,
+	},
+	{
+		name:           "o4 Mini",
+		slugOpenRouter: "openai/o4-mini",
+		slugOpenAI:     openai.ChatModelO4Mini,
+		totalContext:   128000,
+		maxOutput:      65536,
+		inputPrice:     110,
+		outputPrice:    440,
+		requireOwnKey:  true,
+	},
+}
+
 func (s *ChatServerV2) ListSupportedModels(
 	ctx context.Context,
 	req *chatv2.ListSupportedModelsRequest,
@@ -24,89 +220,43 @@ func (s *ChatServerV2) ListSupportedModels(
 		return nil, err
 	}
 
+	hasOwnAPIKey := strings.TrimSpace(settings.OpenAIAPIKey) != ""
+
 	var models []*chatv2.SupportedModel
-	if strings.TrimSpace(settings.OpenAIAPIKey) == "" {
-		models = []*chatv2.SupportedModel{
-			{
-				Name:         "GPT-4.1",
-				Slug:         "openai/gpt-4.1",
-				TotalContext: 1050000,
-				MaxOutput:    32800,
-				InputPrice:   200,
-				OutputPrice:  800,
-			},
-			{
-				Name:         "GPT-4.1-mini",
-				Slug:         "openai/gpt-4.1-mini",
-				TotalContext: 128000,
-				MaxOutput:    16400,
-				InputPrice:   15,
-				OutputPrice:  60,
-			},
-			{
-				Name:         "Qwen Plus (balanced)",
-				Slug:         "qwen/qwen-plus",
-				TotalContext: 131100,
-				MaxOutput:    8200,
-				InputPrice:   40,
-				OutputPrice:  120,
-			},
-			{
-				Name:         "Qwen Turbo (fast)",
-				Slug:         "qwen/qwen-turbo",
-				TotalContext: 1000000,
-				MaxOutput:    8200,
-				InputPrice:   5,
-				OutputPrice:  20,
-			},
-			{
-				Name:         "Gemini 2.5 Flash (fast)",
-				Slug:         "google/gemini-2.5-flash",
-				TotalContext: 1050000,
-				MaxOutput:    65500,
-				InputPrice:   30,
-				OutputPrice:  250,
-			},
-			{
-				Name:         "Gemini 3 Flash Preview",
-				Slug:         "google/gemini-3-flash-preview",
-				TotalContext: 1050000,
-				MaxOutput:    65500,
-				InputPrice:   50,
-				OutputPrice:  300,
-			},
+	for _, config := range allModels {
+		// Choose the appropriate slug based on whether user has their own API key.
+		//
+		// Some models are only available via OpenRouter; for those, slugOpenAI may be empty.
+		// In that case, keep using the OpenRouter slug to avoid returning an empty model slug.
+		slug := config.slugOpenRouter
+		if hasOwnAPIKey && strings.TrimSpace(config.slugOpenAI) != "" {
+			slug = config.slugOpenAI
 		}
-	} else {
-		models = []*chatv2.SupportedModel{
-			{
-				Name:         "GPT-4.1",
-				Slug:         openai.ChatModelGPT4_1,
-				TotalContext: 1050000,
-				MaxOutput:    32800,
-				InputPrice:   200,
-				OutputPrice:  800,
-			},
-			{
-				Name:         "GPT-4o",
-				Slug:         openai.ChatModelGPT4o,
-				TotalContext: 128000,
-				MaxOutput:    16400,
-				InputPrice:   250,
-				OutputPrice:  1000,
-			},
-			{
-				Name:         "GPT-4.1-mini",
-				Slug:         openai.ChatModelGPT4_1Mini,
-				TotalContext: 128000,
-				MaxOutput:    16400,
-				InputPrice:   15,
-				OutputPrice:  60,
-			},
-			// TODO: add user custom models
+
+		model := &chatv2.SupportedModel{
+			Name:         config.name,
+			Slug:         slug,
+			TotalContext: config.totalContext,
+			MaxOutput:    config.maxOutput,
+			InputPrice:   config.inputPrice,
+			OutputPrice:  config.outputPrice,
 		}
+
+		// If model requires own key but user hasn't provided one, mark as disabled
+		if config.requireOwnKey && !hasOwnAPIKey {
+			model.Disabled = true
+			model.DisabledReason = stringPtr("Requires your own OpenAI API key. Configure it in Settings.")
+		}
+
+		models = append(models, model)
 	}
 
 	return &chatv2.ListSupportedModelsResponse{
 		Models: models,
 	}, nil
 }
+
+// stringPtr returns a pointer to the given string
+func stringPtr(s string) *string {
+	return &s
+}

internal/services/system_prompt_debug.tmpl

@@ -1,9 +1,13 @@
 You are PaperDebugger, a large language model tweaked by PaperDebugger Inc.
 
+## tool_call_limit
+You have a maximum of 20 tool calls per conversation turn. Please plan your tool usage carefully and avoid unnecessary tool calls.
+
 ## selected_text
 The user may select sentences or paragraphs of LaTeX content for revision or ask questions along with the selected text.
 If the user asks questions, just answer the question.
-If the user requests to revise the selected text, wrap the revised text in triple backticks.
+If the user requests to revise the selected text, you MUST include a separate block where the revised text is wrapped inside a `<PaperDebugger>` tag, like `<PaperDebugger>...revised text...</PaperDebugger>`.
+The content inside `<PaperDebugger>` MUST be only the revised text (no explanations, no markdown formatting, no surrounding backticks); any explanations should be placed outside of the `<PaperDebugger>` block.
 
 {{ if .ProjectInstructions }}## project_instructions, please follow the project's instructions strictly
 {{ .ProjectInstructions }}{{ end }}

internal/services/system_prompt_default.tmpl

@@ -1,5 +1,8 @@
 You are PaperDebugger, a large language model tweaked by PaperDebugger Inc.
 
+## tool_call_limit
+You have a maximum of 20 tool calls per conversation turn. Please plan your tool usage carefully and avoid unnecessary tool calls.
+
 ## selected_text
 The user may select sentences or paragraphs of LaTeX content for revision. Your task is to revise the selected text according to the user's instructions.