v2.1.3: fix Crof mimo-v2.5-pro reasoning_content token exhaustion

- Strip reasoning_content from proxy output (Codex doesn't use it)
- Force max_tokens=64000 minimum for openai-compat providers
- Prevents models that emit large reasoning from running out of tokens

CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## v2.1.3 (2026-05-19)
+
+- **Fixed Crof mimo-v2.5-pro stopping mid-response (finish_reason=length)**
+  - Root cause: model emits 600+ `reasoning_content` SSE chunks that exhaust `max_tokens` before any actual content is generated
+  - Strip `reasoning_content` from proxy output — Codex doesn't use reasoning, avoids wasting output tokens on invisible text
+  - Force `max_tokens` minimum of 64000 for openai-compat providers — gives models room for both reasoning and content
+  - Works for all openai-compat providers (Crof, Z.AI, DeepSeek, OpenRouter, etc.)
+
 ## v2.1.2 (2026-05-19)
 
 - **Fixed Crof.ai and providers stopping after first tool call (root cause: None tool IDs)**

src/codex-launcher-gui

@@ -24,6 +24,11 @@ model_catalog_json = ""
 """
 
 CHANGELOG = [
+    ("2.1.3", "2026-05-19", [
+        "Fixed Crof mimo-v2.5-pro stopping: reasoning_content exhausted all output tokens",
+        "Strip reasoning_content from proxy output — Codex doesn't use it, avoids token waste",
+        "Force max_tokens=64000 minimum for openai-compat providers — gives models room for both reasoning and content",
+    ]),
     ("2.1.2", "2026-05-19", [
         "Fixed Crof.ai and providers stopping after first tool call (root cause: None tool IDs)",
         "Codex sends function_call items with id=None — proxy now matches tool results to calls by position",
@@ -527,7 +532,7 @@ class LauncherWin(Gtk.Window):
         # header row
         hdr = Gtk.Box(spacing=8)
         vbox.pack_start(hdr, False, False, 0)
-        lbl = Gtk.Label(label="<b>Codex Launcher v2.1.2</b>")
+        lbl = Gtk.Label(label="<b>Codex Launcher v2.1.3</b>")
         lbl.set_use_markup(True)
         hdr.pack_start(lbl, False, False, 0)
         changelog_btn = Gtk.Button(label="Changelog")

src/translate-proxy.py

@@ -370,10 +370,6 @@ def oa_resp_to_responses(chat_resp, model, resp_id=None):
     fm = {"stop": "completed", "length": "incomplete", "tool_calls": "completed", "content_filter": "incomplete"}
     status = fm.get(finish, "incomplete")
     outputs = []
-    rc = msg.get("reasoning_content")
-    if rc:
-        outputs.append({"type": "reasoning", "id": uid("rsn"), "status": "completed",
-                        "content": [{"type": "text", "text": rc}]})
     if content:
         outputs.append({"type": "message", "id": uid("msg"), "role": "assistant", "status": "completed",
                         "content": [{"type": "output_text", "text": content, "annotations": []}]})
@@ -447,9 +443,7 @@ def oa_stream_to_sse(chat_stream, model, req_id):
                 yield emit("response.output_text.delta", {"type": "response.function_call_arguments.delta",
                             "delta": fn["arguments"], "item_id": tc_buf[idx]["id"]})
 
-        rc = delta.get("reasoning_content")
+
-        if rc:
-            yield emit("response.reasoning.delta", {"type": "response.reasoning.delta", "delta": rc})
 
     if msg_opened:
         yield emit("response.output_text.done", {"type": "response.output_text.done",
@@ -885,9 +879,10 @@ class Handler(http.server.BaseHTTPRequestHandler):
         if instructions:
             messages.insert(0, {"role": "system", "content": instructions})
         chat_body = {"model": model, "messages": messages}
-        for k in ("temperature", "top_p", "max_output_tokens"):
+        for k in ("temperature", "top_p"):
             if k in body:
-                chat_body["max_tokens" if k == "max_output_tokens" else k] = body[k]
+                chat_body[k] = body[k]
+        chat_body["max_tokens"] = max(body.get("max_output_tokens", 0), 64000)
         tools = oa_convert_tools(body.get("tools"))
         if tools:
             chat_body["tools"] = tools