@fudrouter/fsrouter 0.6.62 → 0.6.63

package/open-sse/handlers/chatCore/nonStreamingHandler.js

@@ -0,0 +1,300 @@
+import { FORMATS } from "../../translator/formats.js";
+import { needsTranslation } from "../../translator/index.js";
+import { ollamaBodyToOpenAI } from "../../translator/response/ollama-to-openai.js";
+import { addBufferToUsage, filterUsageForFormat } from "../../utils/usageTracking.js";
+import { createErrorResult } from "../../utils/error.js";
+import { HTTP_STATUS } from "../../config/runtimeConfig.js";
+import { parseSSEToOpenAIResponse } from "./sseToJsonHandler.js";
+import { buildRequestDetail, extractRequestConfig, extractUsageFromResponse, saveUsageStats } from "./requestDetail.js";
+import { appendRequestLog, saveRequestDetail } from '../../../dist/lib/usageDb.js';
+import { decloakToolNames } from "../../utils/claudeCloaking.js";
+
+/**
+ * Translate non-streaming response body from provider format → OpenAI format.
+ */
+export function translateNonStreamingResponse(responseBody, targetFormat, sourceFormat) {
+  if (targetFormat === sourceFormat || targetFormat === FORMATS.OPENAI) return responseBody;
+
+  // Gemini / Antigravity
+  if (targetFormat === FORMATS.GEMINI || targetFormat === FORMATS.ANTIGRAVITY || targetFormat === FORMATS.GEMINI_CLI || targetFormat === FORMATS.VERTEX) {
+    const response = responseBody.response || responseBody;
+    if (!response?.candidates?.[0]) return responseBody;
+
+    const candidate = response.candidates[0];
+    const content = candidate.content;
+    const usage = response.usageMetadata || responseBody.usageMetadata;
+    let textContent = "", reasoningContent = "";
+    const toolCalls = [];
+
+    if (content?.parts) {
+      for (const part of content.parts) {
+        if (part.thought === true && part.text) reasoningContent += part.text;
+        else if (part.text !== undefined) textContent += part.text;
+        if (part.functionCall) {
+          toolCalls.push({
+            id: `call_${part.functionCall.name}_${Date.now()}_${toolCalls.length}`,
+            type: "function",
+            function: { name: part.functionCall.name, arguments: JSON.stringify(part.functionCall.args || {}) }
+          });
+        }
+      }
+    }
+
+    const message = { role: "assistant" };
+    if (textContent) message.content = textContent;
+    if (reasoningContent) message.reasoning_content = reasoningContent;
+    if (toolCalls.length > 0) message.tool_calls = toolCalls;
+    if (!message.content && !message.tool_calls) message.content = "";
+
+    let finishReason = (candidate.finishReason || "stop").toLowerCase();
+    if (finishReason === "stop" && toolCalls.length > 0) finishReason = "tool_calls";
+
+    const result = {
+      id: `chatcmpl-${response.responseId || Date.now()}`,
+      object: "chat.completion",
+      created: Math.floor(new Date(response.createTime || Date.now()).getTime() / 1000),
+      model: response.modelVersion || "gemini",
+      choices: [{ index: 0, message, finish_reason: finishReason }]
+    };
+
+    if (usage) {
+      result.usage = {
+        prompt_tokens: (usage.promptTokenCount || 0) + (usage.thoughtsTokenCount || 0),
+        completion_tokens: usage.candidatesTokenCount || 0,
+        total_tokens: usage.totalTokenCount || 0
+      };
+      if (usage.thoughtsTokenCount > 0) {
+        result.usage.completion_tokens_details = { reasoning_tokens: usage.thoughtsTokenCount };
+      }
+    }
+    return result;
+  }
+
+  // Claude
+  if (targetFormat === FORMATS.CLAUDE) {
+    // Always translate a Claude-format body to OpenAI, even if `content` is
+    // missing/null (e.g. M3 with max_tokens:1 spends the budget on thinking
+    // and returns `content: null`). Returning the raw body would leave the
+    // OpenAI client without a `choices` array and surface as a UI test error.
+    if (responseBody.content && !Array.isArray(responseBody.content)) return responseBody;
+
+    let textContent = "", thinkingContent = "";
+    const toolCalls = [];
+
+    for (const block of (responseBody.content || [])) {
+      if (block.type === "text") {
+        // Strip markdown code block markers (e.g. kimi wraps JSON in ```json...```)
+        const raw = block.text ?? "";
+        const text = raw.replace(/^\s*```\s*json\s*\n?/i, "").replace(/\n?\s*```\s*$/i, "");
+        textContent += text;
+      } else if (block.type === "thinking") thinkingContent += block.thinking || "";
+      else if (block.type === "tool_use") {
+        toolCalls.push({ id: block.id, type: "function", function: { name: block.name, arguments: JSON.stringify(block.input || {}) } });
+      }
+    }
+
+    const message = { role: "assistant" };
+    if (textContent) message.content = textContent;
+    if (thinkingContent) message.reasoning_content = thinkingContent;
+    if (toolCalls.length > 0) message.tool_calls = toolCalls;
+    if (!message.content && !message.tool_calls) message.content = "";
+
+    let finishReason = responseBody.stop_reason || "stop";
+    if (finishReason === "end_turn") finishReason = "stop";
+    if (finishReason === "tool_use") finishReason = "tool_calls";
+
+    const result = {
+      id: `chatcmpl-${responseBody.id || Date.now()}`,
+      object: "chat.completion",
+      created: Math.floor(Date.now() / 1000),
+      model: responseBody.model || "claude",
+      choices: [{ index: 0, message, finish_reason: finishReason }]
+    };
+
+    if (responseBody.usage) {
+      result.usage = {
+        prompt_tokens: responseBody.usage.input_tokens || 0,
+        completion_tokens: responseBody.usage.output_tokens || 0,
+        total_tokens: (responseBody.usage.input_tokens || 0) + (responseBody.usage.output_tokens || 0)
+      };
+    }
+    return result;
+  }
+
+  // Ollama
+  if (targetFormat === FORMATS.OLLAMA) {
+    return ollamaBodyToOpenAI(responseBody);
+  }
+
+  return responseBody;
+}
+
+/**
+ * Handle non-streaming response from provider.
+ */
+export async function handleNonStreamingResponse({ providerResponse, provider, model, sourceFormat, targetFormat, body, stream, translatedBody, finalBody, requestStartTime, connectionId, apiKey, clientRawRequest, onRequestSuccess, reqLogger, toolNameMap, trackDone, appendLog }) {
+  trackDone();
+  const contentType = providerResponse.headers.get("content-type") || "";
+  let responseBody;
+
+  if (contentType.includes("text/event-stream")) {
+    const sseText = await providerResponse.text();
+    const parsed = parseSSEToOpenAIResponse(sseText, model);
+    if (!parsed) {
+      appendLog({ status: `FAILED ${HTTP_STATUS.BAD_GATEWAY}` });
+      return createErrorResult(HTTP_STATUS.BAD_GATEWAY, "Invalid SSE response for non-streaming request");
+    }
+    responseBody = parsed;
+  } else {
+    try {
+      responseBody = await providerResponse.json();
+    } catch (err) {
+      appendLog({ status: `FAILED ${HTTP_STATUS.BAD_GATEWAY}` });
+      console.error(`[ChatCore] Failed to parse JSON from ${provider}:`, err.message);
+      return createErrorResult(HTTP_STATUS.BAD_GATEWAY, `Invalid JSON response from ${provider}`);
+    }
+  }
+
+  reqLogger.logProviderResponse(providerResponse.status, providerResponse.statusText, providerResponse.headers, responseBody);
+  if (onRequestSuccess) await onRequestSuccess();
+
+  // Decloak tool_use names once on raw Claude body, before any translation (INPUT side)
+  responseBody = decloakToolNames(responseBody, toolNameMap);
+
+  const usage = extractUsageFromResponse(responseBody);
+  appendLog({ tokens: usage, status: "200 OK" });
+  saveUsageStats({ provider, model, tokens: usage, connectionId, apiKey, endpoint: clientRawRequest?.endpoint });
+
+  const translatedResponse = needsTranslation(targetFormat, sourceFormat)
+    ? translateNonStreamingResponse(responseBody, targetFormat, sourceFormat)
+    : responseBody;
+
+  // Fix finish_reason for tool_calls: some providers return non-standard values (e.g. "other")
+  if (translatedResponse?.choices?.[0]) {
+    const choice = translatedResponse.choices[0];
+    const msg = choice.message;
+    const hasToolCalls = Array.isArray(msg?.tool_calls) && msg.tool_calls.length > 0;
+    if (hasToolCalls && choice.finish_reason !== "tool_calls") {
+      choice.finish_reason = "tool_calls";
+    }
+  }
+
+  // Ensure OpenAI-required fields
+  if (!translatedResponse.object) translatedResponse.object = "chat.completion";
+  if (!translatedResponse.created) translatedResponse.created = Math.floor(Date.now() / 1000);
+
+  // Strip Azure-specific fields
+  delete translatedResponse.prompt_filter_results;
+  if (translatedResponse?.choices) {
+    for (const choice of translatedResponse.choices) delete choice.content_filter_results;
+  }
+
+  if (translatedResponse?.usage) {
+    translatedResponse.usage = filterUsageForFormat(addBufferToUsage(translatedResponse.usage), sourceFormat);
+  }
+
+  // Strip reasoning_content — some clients (e.g. Firecrawl AI SDK) have JSON parsers that
+  // break on this non-standard field, even though OpenAI allows it in extensions.
+  if (translatedResponse?.choices && sourceFormat !== FORMATS.CLAUDE) {
+    for (const choice of translatedResponse.choices) {
+      if (choice?.message) delete choice.message.reasoning_content;
+    }
+  }
+
+  // If client sent Anthropic Messages format (/v1/messages), convert response back to
+  // Anthropic format. 9router always works in OpenAI format internally; this step
+  // is the reverse-translation for Claude Code / Anthropic SDK clients.
+  if (sourceFormat === FORMATS.CLAUDE) {
+    const choice = translatedResponse?.choices?.[0];
+    const msg = choice?.message || {};
+    const contentBlocks = [];
+    if (msg.reasoning_content) contentBlocks.push({ type: "thinking", thinking: msg.reasoning_content });
+    if (msg.content) contentBlocks.push({ type: "text", text: msg.content });
+    if (Array.isArray(msg.tool_calls)) {
+      for (const tc of msg.tool_calls) {
+        let input = {};
+        try { input = JSON.parse(tc.function?.arguments || "{}"); } catch {}
+        contentBlocks.push({ type: "tool_use", id: tc.id, name: tc.function?.name, input });
+      }
+    }
+    // Claude Code requires at least one text or tool_use block — cannot be only thinking
+    const hasTextOrTool = contentBlocks.some(b => b.type === "text" || b.type === "tool_use");
+    if (!hasTextOrTool) contentBlocks.push({ type: "text", text: "" });
+
+    let stopReason = "end_turn";
+    const fr = choice?.finish_reason;
+    if (fr === "tool_calls") stopReason = "tool_use";
+    else if (fr === "length") stopReason = "max_tokens";
+
+    const anthropicResponse = {
+      id: translatedResponse.id || `msg_${Date.now()}`,
+      type: "message",
+      role: "assistant",
+      content: contentBlocks.length ? contentBlocks : [{ type: "text", text: "" }],
+      model: translatedResponse.model || model,
+      stop_reason: stopReason,
+      stop_sequence: null,
+    };
+    if (translatedResponse.usage) {
+      anthropicResponse.usage = {
+        input_tokens: translatedResponse.usage.prompt_tokens || 0,
+        output_tokens: translatedResponse.usage.completion_tokens || 0,
+      };
+    } else {
+      anthropicResponse.usage = { input_tokens: 0, output_tokens: 0 };
+    }
+
+    reqLogger.logConvertedResponse(anthropicResponse);
+    const totalLatency = Date.now() - requestStartTime;
+    saveRequestDetail(buildRequestDetail({
+      provider, model, connectionId,
+      latency: { ttft: totalLatency, total: totalLatency },
+      tokens: usage || { prompt_tokens: 0, completion_tokens: 0 },
+      request: extractRequestConfig(body, stream),
+      providerRequest: finalBody || translatedBody || null,
+      providerResponse: responseBody || null,
+      response: {
+        content: msg.content || null,
+        thinking: msg.reasoning_content || null,
+        finish_reason: choice?.finish_reason || "unknown"
+      },
+      status: "success"
+    }, { endpoint: clientRawRequest?.endpoint || null })).catch(err => {
+      console.error("[RequestDetail] Failed to save:", err.message);
+    });
+
+    return {
+      success: true,
+      response: new Response(JSON.stringify(anthropicResponse), {
+        headers: { "Content-Type": "application/json", "Access-Control-Allow-Origin": "*" }
+      })
+    };
+  }
+
+  reqLogger.logConvertedResponse(translatedResponse);
+
+  const totalLatency = Date.now() - requestStartTime;
+  saveRequestDetail(buildRequestDetail({
+    provider, model, connectionId,
+    latency: { ttft: totalLatency, total: totalLatency },
+    tokens: usage || { prompt_tokens: 0, completion_tokens: 0 },
+    request: extractRequestConfig(body, stream),
+    providerRequest: finalBody || translatedBody || null,
+    providerResponse: responseBody || null,
+    response: {
+      content: translatedResponse?.choices?.[0]?.message?.content || translatedResponse?.content || null,
+      thinking: translatedResponse?.choices?.[0]?.message?.reasoning_content || translatedResponse?.reasoning_content || null,
+      finish_reason: translatedResponse?.choices?.[0]?.finish_reason || "unknown"
+    },
+    status: "success"
+  }, { endpoint: clientRawRequest?.endpoint || null })).catch(err => {
+    console.error("[RequestDetail] Failed to save:", err.message);
+  });
+
+  return {
+    success: true,
+    response: new Response(JSON.stringify(translatedResponse), {
+      headers: { "Content-Type": "application/json", "Access-Control-Allow-Origin": "*" }
+    })
+  };
+}

package/open-sse/handlers/chatCore/requestDetail.js

@@ -0,0 +1,102 @@
+import { saveRequestUsage, appendRequestLog, saveRequestDetail } from '../../../dist/lib/usageDb.js';
+import { COLORS } from "../../utils/stream.js";
+
+const OPTIONAL_PARAMS = [
+  "temperature", "top_p", "top_k",
+  "max_tokens", "max_completion_tokens",
+  "thinking", "reasoning", "enable_thinking",
+  "presence_penalty", "frequency_penalty",
+  "seed", "stop", "tools", "tool_choice",
+  "response_format", "prediction", "store", "metadata",
+  "n", "logprobs", "top_logprobs", "logit_bias",
+  "user", "parallel_tool_calls"
+];
+
+export function extractRequestConfig(body, stream) {
+  const config = { messages: body.messages || [], model: body.model, stream };
+  for (const param of OPTIONAL_PARAMS) {
+    if (body[param] !== undefined) config[param] = body[param];
+  }
+  return config;
+}
+
+export function extractUsageFromResponse(responseBody) {
+  if (!responseBody || typeof responseBody !== "object") return null;
+
+  // Claude format
+  if (responseBody.usage?.input_tokens !== undefined) {
+    return {
+      prompt_tokens: responseBody.usage.input_tokens || 0,
+      completion_tokens: responseBody.usage.output_tokens || 0,
+      cache_read_input_tokens: responseBody.usage.cache_read_input_tokens,
+      cache_creation_input_tokens: responseBody.usage.cache_creation_input_tokens
+    };
+  }
+
+  // OpenAI format
+  if (responseBody.usage?.prompt_tokens !== undefined) {
+    return {
+      prompt_tokens: responseBody.usage.prompt_tokens || 0,
+      completion_tokens: responseBody.usage.completion_tokens || 0,
+      cached_tokens: responseBody.usage.prompt_tokens_details?.cached_tokens,
+      reasoning_tokens: responseBody.usage.completion_tokens_details?.reasoning_tokens
+    };
+  }
+
+  // Gemini format
+  if (responseBody.usageMetadata) {
+    return {
+      prompt_tokens: responseBody.usageMetadata.promptTokenCount || 0,
+      completion_tokens: responseBody.usageMetadata.candidatesTokenCount || 0,
+      reasoning_tokens: responseBody.usageMetadata.thoughtsTokenCount
+    };
+  }
+
+  return null;
+}
+
+export function buildRequestDetail(base, overrides = {}) {
+  return {
+    provider: base.provider || "unknown",
+    model: base.model || "unknown",
+    connectionId: base.connectionId || undefined,
+    timestamp: new Date().toISOString(),
+    latency: base.latency || { ttft: 0, total: 0 },
+    tokens: base.tokens || { prompt_tokens: 0, completion_tokens: 0 },
+    request: base.request,
+    providerRequest: base.providerRequest || null,
+    providerResponse: base.providerResponse || null,
+    response: base.response || {},
+    status: base.status || "success",
+    ...overrides
+  };
+}
+
+export function saveUsageStats({ provider, model, tokens, connectionId, apiKey, endpoint, label = "USAGE" }) {
+  if (!tokens || typeof tokens !== "object") return;
+
+  const inTokens = tokens.input_tokens ?? tokens.prompt_tokens ?? 0;
+  const outTokens = tokens.output_tokens ?? tokens.completion_tokens ?? 0;
+
+  if (inTokens === 0 && outTokens === 0) return;
+
+  const time = new Date().toLocaleTimeString("en-US", { hour12: false, hour: "2-digit", minute: "2-digit", second: "2-digit" });
+  const accountSuffix = connectionId ? ` | account=${connectionId.slice(0, 8)}...` : "";
+  console.log(`${COLORS.green}[${time}] 📊 [${label}] ${provider.toUpperCase()} | in=${inTokens} | out=${outTokens}${accountSuffix}${COLORS.reset}`);
+
+  // Normalize to OpenAI token shape for storage
+  const normalized = {
+    prompt_tokens: tokens.prompt_tokens ?? tokens.input_tokens ?? 0,
+    completion_tokens: tokens.completion_tokens ?? tokens.output_tokens ?? 0
+  };
+
+  saveRequestUsage({
+    provider: provider || "unknown",
+    model: model || "unknown",
+    tokens: normalized,
+    timestamp: new Date().toISOString(),
+    connectionId: connectionId || undefined,
+    apiKey: apiKey || undefined,
+    endpoint: endpoint || null
+  }).catch(() => {});
+}

package/open-sse/handlers/chatCore/sseToJsonHandler.js

@@ -0,0 +1,231 @@
+import { convertResponsesStreamToJson } from "../../transformer/streamToJsonConverter.js";
+import { createErrorResult } from "../../utils/error.js";
+import { HTTP_STATUS } from "../../config/runtimeConfig.js";
+import { FORMATS } from "../../translator/formats.js";
+import { buildRequestDetail, extractRequestConfig, saveUsageStats } from "./requestDetail.js";
+import { saveRequestDetail, appendRequestLog } from '../../../dist/lib/usageDb.js';
+
+function textFromResponsesMessageItem(item) {
+  if (!item?.content || !Array.isArray(item.content)) return "";
+  const byType = item.content.find((c) => c.type === "output_text");
+  if (typeof byType?.text === "string") return byType.text;
+  const anyText = item.content.find((c) => typeof c.text === "string");
+  if (typeof anyText?.text === "string") return anyText.text;
+  return "";
+}
+
+/**
+ * Codex / Responses API may emit many alternating reasoning + message items.
+ * Early message blocks often have empty output_text; the user-visible answer is usually in the last non-empty message.
+ */
+function pickAssistantMessageForChatCompletion(output) {
+  if (!Array.isArray(output)) return { msgItem: null, textContent: null };
+  const messages = output.filter((item) => item?.type === "message");
+  if (messages.length === 0) return { msgItem: null, textContent: null };
+  for (let i = messages.length - 1; i >= 0; i--) {
+    const text = textFromResponsesMessageItem(messages[i]);
+    if (text.length > 0) return { msgItem: messages[i], textContent: text };
+  }
+  const last = messages[messages.length - 1];
+  return { msgItem: last, textContent: textFromResponsesMessageItem(last) };
+}
+
+/**
+ * Parse OpenAI-style SSE text into a single chat completion JSON.
+ * Used when provider forces streaming but client wants non-streaming.
+ */
+export function parseSSEToOpenAIResponse(rawSSE, fallbackModel) {
+  const chunks = [];
+
+  for (const line of String(rawSSE || "").split("\n")) {
+    const trimmed = line.trim();
+    if (!trimmed.startsWith("data:")) continue;
+    const payload = trimmed.slice(5).trim();
+    if (!payload || payload === "[DONE]") continue;
+    try { chunks.push(JSON.parse(payload)); } catch { /* ignore malformed lines */ }
+  }
+
+  if (chunks.length === 0) return null;
+
+  const first = chunks[0];
+  const contentParts = [];
+  const reasoningParts = [];
+  const toolCallMap = new Map(); // index -> { id, type, function: { name, arguments } }
+  let finishReason = "stop";
+  let usage = null;
+
+  for (const chunk of chunks) {
+    const choice = chunk?.choices?.[0];
+    const delta = choice?.delta || {};
+    if (typeof delta.content === "string" && delta.content.length > 0) contentParts.push(delta.content);
+    if (typeof delta.reasoning_content === "string" && delta.reasoning_content.length > 0) reasoningParts.push(delta.reasoning_content);
+    if (choice?.finish_reason) finishReason = choice.finish_reason;
+    if (chunk?.usage && typeof chunk.usage === "object") usage = chunk.usage;
+
+    // Accumulate tool_calls from streaming deltas
+    if (Array.isArray(delta.tool_calls)) {
+      for (const tc of delta.tool_calls) {
+        const idx = tc.index ?? 0;
+        if (!toolCallMap.has(idx)) {
+          toolCallMap.set(idx, { id: tc.id || "", type: "function", function: { name: "", arguments: "" } });
+        }
+        const existing = toolCallMap.get(idx);
+        if (tc.id) existing.id = tc.id;
+        if (tc.function?.name) existing.function.name += tc.function.name;
+        if (tc.function?.arguments) existing.function.arguments += tc.function.arguments;
+      }
+    }
+  }
+
+  const message = { role: "assistant", content: contentParts.join("") || (toolCallMap.size > 0 ? null : "") };
+  if (reasoningParts.length > 0) message.reasoning_content = reasoningParts.join("");
+  if (toolCallMap.size > 0) {
+    message.tool_calls = [...toolCallMap.entries()].sort((a, b) => a[0] - b[0]).map(([, tc]) => tc);
+  }
+
+  const result = {
+    id: first.id || `chatcmpl-${Date.now()}`,
+    object: "chat.completion",
+    created: first.created || Math.floor(Date.now() / 1000),
+    model: first.model || fallbackModel || "unknown",
+    choices: [{ index: 0, message, finish_reason: finishReason }]
+  };
+  if (usage) result.usage = usage;
+  return result;
+}
+
+/**
+ * Handle case: provider forced streaming but client wants JSON.
+ * Supports both Codex/Responses API SSE and standard Chat Completions SSE.
+ */
+export async function handleForcedSSEToJson({ providerResponse, sourceFormat, provider, model, body, stream, translatedBody, finalBody, requestStartTime, connectionId, apiKey, clientRawRequest, onRequestSuccess, trackDone, appendLog }) {
+  const contentType = providerResponse.headers.get("content-type") || "";
+  const isSSE = contentType.includes("text/event-stream") || (contentType === "" && provider === "codex");
+  if (!isSSE) return null; // not handled here
+
+  trackDone();
+
+  const ctx = {
+    provider, model, connectionId,
+    request: extractRequestConfig(body, stream),
+    providerRequest: finalBody || translatedBody || null
+  };
+
+  // Codex/Responses API SSE path
+  const isCodexResponsesApi = provider === "codex" || sourceFormat === FORMATS.OPENAI_RESPONSES;
+  if (isCodexResponsesApi) {
+    try {
+      const jsonResponse = await convertResponsesStreamToJson(providerResponse.body);
+      if (onRequestSuccess) await onRequestSuccess();
+
+      const usage = jsonResponse.usage || {};
+      appendLog({ tokens: usage, status: "200 OK" });
+      saveUsageStats({ provider, model, tokens: usage, connectionId, apiKey, endpoint: clientRawRequest?.endpoint });
+
+      const { msgItem, textContent } = pickAssistantMessageForChatCompletion(jsonResponse.output);
+      const totalLatency = Date.now() - requestStartTime;
+
+      saveRequestDetail(buildRequestDetail({
+        ...ctx,
+        latency: { ttft: totalLatency, total: totalLatency },
+        tokens: { prompt_tokens: usage.input_tokens || 0, completion_tokens: usage.output_tokens || 0 },
+        response: { content: textContent, thinking: null, finish_reason: jsonResponse.status || "unknown" },
+        status: "success"
+      }, { endpoint: clientRawRequest?.endpoint || null })).catch(() => {});
+
+      // Client is Responses API → return as-is
+      if (sourceFormat === FORMATS.OPENAI_RESPONSES) {
+        return { success: true, response: new Response(JSON.stringify(jsonResponse), { headers: { "Content-Type": "application/json", "Access-Control-Allow-Origin": "*" } }) };
+      }
+
+      // Build client-format response
+      const inTokens = usage.input_tokens || 0;
+      const outTokens = usage.output_tokens || 0;
+      let finalResp;
+
+      // Extract tool calls from Responses API output (function_call items)
+      const funcCallItems = (jsonResponse.output || []).filter(item => item.type === "function_call");
+      const toolCalls = funcCallItems.map((item, idx) => ({
+        id: item.call_id || `call_${item.name}_${Date.now()}_${idx}`,
+        type: "function",
+        function: {
+          name: item.name,
+          arguments: typeof item.arguments === "string" ? item.arguments : JSON.stringify(item.arguments || {})
+        }
+      }));
+      const hasToolCalls = toolCalls.length > 0;
+
+      if (sourceFormat === FORMATS.ANTIGRAVITY || sourceFormat === FORMATS.GEMINI || sourceFormat === FORMATS.GEMINI_CLI) {
+        finalResp = {
+          response: {
+            candidates: [{ content: { role: "model", parts: [{ text: textContent || "" }] }, finishReason: "STOP", index: 0 }],
+            usageMetadata: { promptTokenCount: inTokens, candidatesTokenCount: outTokens, totalTokenCount: inTokens + outTokens },
+            modelVersion: model,
+            responseId: jsonResponse.id || `resp_${Date.now()}`
+          }
+        };
+      } else {
+        const message = { role: "assistant", content: textContent || (hasToolCalls ? null : "") };
+        if (hasToolCalls) message.tool_calls = toolCalls;
+        const responseDone = jsonResponse.status === "completed" || jsonResponse.status === "done";
+        const finishReason = hasToolCalls ? "tool_calls" : (responseDone ? "stop" : (jsonResponse.status || "stop"));
+        finalResp = {
+          id: jsonResponse.id || `chatcmpl-${Date.now()}`,
+          object: "chat.completion",
+          created: jsonResponse.created_at || Math.floor(Date.now() / 1000),
+          model: jsonResponse.model || model,
+          choices: [{ index: 0, message, finish_reason: finishReason }],
+          usage: { prompt_tokens: inTokens, completion_tokens: outTokens, total_tokens: inTokens + outTokens }
+        };
+      }
+
+      return { success: true, response: new Response(JSON.stringify(finalResp), { headers: { "Content-Type": "application/json", "Access-Control-Allow-Origin": "*" } }) };
+    } catch (err) {
+      console.error("[ChatCore] Responses API SSE→JSON failed:", err);
+      return createErrorResult(HTTP_STATUS.BAD_GATEWAY, "Failed to convert streaming response to JSON");
+    }
+  }
+
+  // Standard Chat Completions SSE path
+  try {
+    const sseText = await providerResponse.text();
+    const parsed = parseSSEToOpenAIResponse(sseText, model);
+    if (!parsed) return createErrorResult(HTTP_STATUS.BAD_GATEWAY, "Invalid SSE response for non-streaming request");
+
+    if (onRequestSuccess) await onRequestSuccess();
+
+    const usage = parsed.usage || {};
+    appendLog({ tokens: usage, status: "200 OK" });
+    saveUsageStats({ provider, model, tokens: usage, connectionId, apiKey, endpoint: clientRawRequest?.endpoint });
+
+    const totalLatency = Date.now() - requestStartTime;
+    saveRequestDetail(buildRequestDetail({
+      ...ctx,
+      latency: { ttft: totalLatency, total: totalLatency },
+      tokens: usage,
+      response: {
+        content: parsed.choices?.[0]?.message?.content || null,
+        thinking: parsed.choices?.[0]?.message?.reasoning_content || null,
+        finish_reason: parsed.choices?.[0]?.finish_reason || "unknown"
+      },
+      status: "success"
+    }, { endpoint: clientRawRequest?.endpoint || null })).catch(() => {});
+
+    // Strip reasoning_content only when content is non-empty.
+    // When content is empty (e.g. thinking models that used all tokens for reasoning),
+    // reasoning_content is the only useful output and must be preserved.
+    // Previously this was unconditional, which broke Qwen3.5, Claude extended thinking, etc.
+    if (parsed?.choices) {
+      for (const choice of parsed.choices) {
+        if (choice?.message?.reasoning_content && choice.message.content) {
+          delete choice.message.reasoning_content;
+        }
+      }
+    }
+
+    return { success: true, response: new Response(JSON.stringify(parsed), { headers: { "Content-Type": "application/json", "Access-Control-Allow-Origin": "*" } }) };
+  } catch (err) {
+    console.error("[ChatCore] Chat Completions SSE→JSON failed:", err);
+    return createErrorResult(HTTP_STATUS.BAD_GATEWAY, "Failed to convert streaming response to JSON");
+  }
+}