@oneciel-ai/claude-any 0.1.38 → 0.1.42

package/README.md

@@ -48,7 +48,7 @@ arguments through unchanged.
 
 Credits: One Ciel LLC
 
-Current version: `0.1.38`
+Current version: `0.1.42`
 
 ## Why This Exists
 
@@ -381,6 +381,30 @@ steps under that larger model's supervision.
 
 ## Changelog
 
+### 0.1.42
+
+- **Live stream progress**: the statusline now updates streamed upstream output
+  progress with formatted input/output token estimates and chunk counts.
+
+### 0.1.41
+
+- **Statusline formatting**: upstream token counts now use thousands separators
+  and a space before `tok`, for example `27,501 tok`.
+
+### 0.1.40
+
+- **RPM 0 is preserved**: setting `rate_limit_rpm=0` now stores an explicit
+  unlimited mode instead of falling back to the provider default.
+
+### 0.1.39
+
+- **Menu input fixes**: restores terminal line/echo mode before text or number
+  prompts, so typed numeric values are visible in the prelaunch UI.
+- **Safer numeric validation**: invalid numeric option input now shows an
+  inline message instead of crashing the menu.
+- **Preset visibility**: applied presets report the effective context, reserve,
+  output, and timeout values.
+
 ### 0.1.38
 
 - **User-selected context windows**: removes the NVIDIA hosted 32K safety cap.

package/claude_any.py

@@ -85,7 +85,7 @@ PROVIDER_LABELS = {
     "self-hosted-nim": "Self Hosted NIM",
 }
 APP_NAME = "Claude Any"
-VERSION = "0.1.38"
+VERSION = "0.1.42"
 CREDITS = "Credits: One Ciel LLC"
 
 LOG_LEVELS = {"SILENT": 0, "ERROR": 1, "WARN": 2, "INFO": 3, "DEBUG": 4, "TRACE": 5}
@@ -1339,7 +1339,22 @@ def main():
                 tokens = activity.get("tokens")
                 rpm_text += f" | upstream {age:.0f}s"
                 if tokens:
-                    rpm_text += f" {tokens}tok"
+                    try:
+                        rpm_text += f" {int(tokens):,} tok"
+                    except Exception:
+                        rpm_text += f" {tokens} tok"
+                output_tokens = activity.get("output_tokens")
+                if output_tokens:
+                    try:
+                        rpm_text += f" -> {int(output_tokens):,} tok"
+                    except Exception:
+                        rpm_text += f" -> {output_tokens} tok"
+                chunks = activity.get("chunks")
+                if chunks:
+                    try:
+                        rpm_text += f" ({int(chunks):,} chunks)"
+                    except Exception:
+                        rpm_text += f" ({chunks} chunks)"
             elif event in ("success", "error"):
                 rpm_text += f" | {event} {age:.0f}s"
     print(f"{left} | {color(rpm_text)}")
@@ -2207,8 +2222,10 @@ def router_rate_limit_recent(timestamps: Any, now: float, window: float, *, incl
 
 def router_rate_limit_usage(provider: str, pcfg: dict[str, Any], model: str | None = None) -> tuple[int, int | None]:
     rpm = router_rate_limit_effective_rpm(provider, pcfg, model)
-    if rpm is None:
-        return 0, None
+    if rpm is None:
+        return 0, None
+    if rpm == 0:
+        return 0, 0
     key = router_rate_limit_key(provider, pcfg, model)
     now = time.time()
     try:
@@ -4697,6 +4714,8 @@ def stream_openai_chat_to_anthropic_sse(
     source_body: dict[str, Any] | None = None,
     start_index: int = 0,
     word_chunking: bool = False,
+    input_tokens: int | None = None,
+    input_bytes: int | None = None,
 ) -> None:
     next_content_index = start_index
     text_started = False
@@ -4709,6 +4728,8 @@ def stream_openai_chat_to_anthropic_sse(
     tool_fragments: dict[int, dict[str, Any]] = {}
     output_tokens = 0
     finish_reason = "stop"
+    chunks_seen = 0
+    last_activity_update = 0.0
 
     def emit(event_name: str, payload: dict[str, Any]) -> None:
         handler.wfile.write(f"event: {event_name}\ndata: {json.dumps(payload, ensure_ascii=False)}\n\n".encode())
@@ -4736,8 +4757,27 @@ def stream_openai_chat_to_anthropic_sse(
             {"type": "content_block_delta", "index": idx, "delta": {"type": "text_delta", "text": text}},
         )
 
+    def update_stream_activity(force: bool = False) -> None:
+        nonlocal last_activity_update
+        now = time.time()
+        if not force and now - last_activity_update < 0.5:
+            return
+        last_activity_update = now
+        estimated_output = output_tokens or max(0, len(text_so_far) // 4)
+        write_router_activity(
+            "request",
+            provider,
+            model,
+            tokens=input_tokens,
+            bytes=input_bytes,
+            output_tokens=estimated_output,
+            chunks=chunks_seen,
+            stream=True,
+        )
+
     try:
         for raw_line in resp:
+            chunks_seen += 1
             line = raw_line.decode("utf-8", errors="ignore").strip()
             if not line or line.startswith(":"):
                 continue
@@ -4789,6 +4829,7 @@ def stream_openai_chat_to_anthropic_sse(
                     emit_text_delta(to_flush)
                 else:
                     emit_text_delta(text_chunk)
+                update_stream_activity()
             for call in delta.get("tool_calls") or []:
                 if not isinstance(call, dict):
                     continue
@@ -4804,6 +4845,8 @@ def stream_openai_chat_to_anthropic_sse(
                     slot["name"] += str(fn.get("name"))
                 if fn.get("arguments"):
                     slot["arguments"] += str(fn.get("arguments"))
+                update_stream_activity()
+        update_stream_activity(force=True)
         if word_chunking and text_buffer:
             to_flush, text_buffer = _split_word_buffer(text_buffer, force=True)
             emit_text_delta(to_flush)
@@ -5108,6 +5151,8 @@ def forward_openai_compatible_chat(handler: BaseHTTPRequestHandler, provider: st
                 model,
                 emit_retry_notice,
             )
+            req_tokens = estimate_tokens(req_body)
+            req_bytes = len(json.dumps(req_body, ensure_ascii=False).encode("utf-8"))
             stream_openai_chat_to_anthropic_sse(
                 handler,
                 resp,
@@ -5116,8 +5161,10 @@ def forward_openai_compatible_chat(handler: BaseHTTPRequestHandler, provider: st
                 source_body=body,
                 start_index=index,
                 word_chunking=bool(pcfg.get("stream_word_chunking", False)),
+                input_tokens=req_tokens,
+                input_bytes=req_bytes,
             )
-            write_router_activity("success", provider, model, tokens=estimate_tokens(req_body), bytes=len(json.dumps(req_body, ensure_ascii=False).encode("utf-8")), stream=True)
+            write_router_activity("success", provider, model, tokens=req_tokens, bytes=req_bytes, stream=True)
         except RuntimeError as exc:
             msg = str(exc)
             write_anthropic_stream_blocks(handler, [{"type": "text", "text": f"Upstream error: {msg}"}], index)
@@ -6282,16 +6329,38 @@ def apply_llm_preset_to_provider(provider: str, pcfg: dict[str, Any], preset_id:
         f"{ui_text('apply_preset', lang)}: {label}",
         f"Provider: {provider}; {ui_text('model_family', lang)}: {model_family_text(family, lang)}",
     ]
-    if provider in ("vllm", "self-hosted-nim"):
-        server_limit = upstream_model_context_limit(provider, pcfg)
-        if server_limit:
-            lines.append(f"Server max_model_len: {server_limit}")
-            if preset_id in ("long-context-65k", "large-output") and server_limit < 65536:
-                lines.append("Long-context preset requires restarting the server with --max-model-len 65536 or higher.")
-                lines.append("Client settings were capped to the server-reported context length.")
-        elif preset_id in ("long-context-65k", "large-output"):
-            lines.append("Could not verify server max_model_len; vLLM/NIM must be started with a matching context limit.")
-    return lines
+    if provider in ("vllm", "nvidia-hosted", "self-hosted-nim"):
+        server_limit = upstream_model_context_limit(provider, pcfg)
+        if server_limit:
+            lines.append(f"Server max_model_len: {server_limit}")
+            if preset_id in ("long-context-65k", "large-output") and server_limit < 65536:
+                lines.append("Long-context preset requires restarting the server with --max-model-len 65536 or higher.")
+                lines.append("Client settings were capped to the server-reported context length.")
+        elif preset_id in ("long-context-65k", "large-output"):
+            lines.append("Could not verify server max_model_len; vLLM/NIM must be started with a matching context limit.")
+    if provider in ("vllm", "nvidia-hosted", "self-hosted-nim"):
+        lines.append(
+            "Applied options: "
+            f"context_window={pcfg.get('context_window', 'default')}, "
+            f"reserve={pcfg.get('context_reserve_tokens', 'default')}, "
+            f"max_output_tokens={pcfg.get('max_output_tokens', 'default')}, "
+            f"timeout={pcfg.get('request_timeout_ms', 'default')}ms"
+        )
+    elif provider in ("ollama", "ollama-cloud"):
+        opts = ollama_extra_options(pcfg)
+        lines.append(
+            "Applied options: "
+            f"num_ctx={ollama_num_ctx_status(pcfg)}, "
+            f"num_predict={opts.get('num_predict', 'default')}, "
+            f"timeout={pcfg.get('request_timeout_ms', 'default')}ms"
+        )
+    elif provider == "anthropic":
+        lines.append(
+            "Applied options: "
+            f"max_output_tokens={pcfg.get('max_output_tokens', 'default')}, "
+            f"timeout={pcfg.get('request_timeout_ms', 'default')}ms"
+        )
+    return lines
 
 
 def apply_llm_preset_config(provider: str, preset_id: str) -> list[str]:
@@ -6489,9 +6558,9 @@ def llm_option_panel_rows(provider: str, pcfg: dict[str, Any], lang: str | None
         add("Rate limit RPM", "rate_limit_rpm", pcfg.get("rate_limit_rpm", 40))
         add("Rate limit status", "rate_limit_status", "on" if bool(pcfg.get("rate_limit_status", True)) else "off")
     else:
-        if provider in ("vllm", "self-hosted-nim"):
-            add("Context window", "context_window", pcfg.get("context_window", "default"))
-            add("Context reserve", "context_reserve_tokens", pcfg.get("context_reserve_tokens", "default"))
+        if provider in ("vllm", "nvidia-hosted", "self-hosted-nim"):
+            add("Context window", "context_window", pcfg.get("context_window", "default"))
+            add("Context reserve", "context_reserve_tokens", pcfg.get("context_reserve_tokens", "default"))
         add("Max output tokens", "max_output_tokens", pcfg.get("max_output_tokens", "default"))
         if provider in ("vllm", "nvidia-hosted", "self-hosted-nim"):
             add("Timeout ms", "request_timeout_ms", pcfg.get("request_timeout_ms", "default"))
@@ -6532,13 +6601,38 @@ def llm_option_prompt_default(provider: str, pcfg: dict[str, Any], key: str) ->
     return "" if value is None else str(value)
 
 
-def set_llm_option_config(provider: str, key: str, raw_value: str) -> list[str]:
+def set_llm_option_config(provider: str, key: str, raw_value: str) -> list[str]:
     cfg = load_config()
     pcfg = cfg["providers"][provider]
-    value = raw_value.strip()
-    if not value:
-        return ["Option unchanged."]
-    clear_words = ("default", "unset", "none", "null")
+    value = raw_value.strip()
+    if not value:
+        return ["Option unchanged."]
+    numeric_keys = {
+        "context_window",
+        "context",
+        "max_model_len",
+        "context_reserve_tokens",
+        "reserve",
+        "max_output_tokens",
+        "max_tokens",
+        "maxtoken",
+        "max_token",
+        "num_ctx_min",
+        "num_ctx_max",
+        "num_predict",
+        "timeout",
+        "timeout_ms",
+        "request_timeout",
+        "request_timeout_ms",
+        "rate_limit",
+        "rate_limit_rpm",
+        "rpm",
+        "top_k",
+    }
+    if key in numeric_keys and value.lower() not in ("default", "unset", "none", "null", "0"):
+        if not re.fullmatch(r"\d+", value):
+            return [f"{key}: enter digits only, or use default/unset to clear."]
+    clear_words = ("default", "unset", "none", "null")
     token = f"unset:{key}" if value.lower() in clear_words else f"{key}={value}"
     if provider in ("ollama", "ollama-cloud"):
         apply_ollama_option(pcfg, token)
@@ -6616,11 +6710,11 @@ def apply_provider_option(provider: str, pcfg: dict[str, Any], token: str) -> No
             raise SystemExit("timeout must be a positive integer; values above 10000 are treated as milliseconds")
         pcfg["request_timeout_ms"] = fixed if key.endswith("_ms") or fixed > 10000 else fixed * 1000
         return
-    if key in ("rate_limit", "rate_limit_rpm", "rpm"):
-        fixed = positive_int(value)
-        if value in (0, "0", False, None):
-            pcfg.pop("rate_limit_rpm", None)
-            return
+    if key in ("rate_limit", "rate_limit_rpm", "rpm"):
+        fixed = positive_int(value)
+        if value in (0, "0", False, None):
+            pcfg["rate_limit_rpm"] = 0
+            return
         if not fixed:
             raise SystemExit("rate_limit_rpm must be a positive integer, or 0/unset to disable")
         pcfg["rate_limit_rpm"] = fixed
@@ -8116,14 +8210,16 @@ def render_prelaunch_screen(
     return False
 
 
-def prompt_menu_value(prompt: str, default: str = "", secret: bool = False) -> str:
-    label = f"{prompt}"
-    if default:
-        label += f" [{default}]"
-    label += ": "
-    if sys.stdout.isatty():
-        sys.stdout.write("\033[?25h")
-        sys.stdout.flush()
+def prompt_menu_value(prompt: str, default: str = "", secret: bool = False, restore_tty: Callable[[], None] | None = None, raw_tty: Callable[[], None] | None = None) -> str:
+    label = f"{prompt}"
+    if default:
+        label += f" [{default}]"
+    label += ": "
+    if restore_tty:
+        restore_tty()
+    if sys.stdout.isatty():
+        sys.stdout.write("\033[?25h")
+        sys.stdout.flush()
     sys.stdout.write("\n" + ansi(label, "1;38;5;208"))
     sys.stdout.flush()
     try:
@@ -8131,12 +8227,14 @@ def prompt_menu_value(prompt: str, default: str = "", secret: bool = False) -> s
             value = getpass.getpass("")
         else:
             value = input()
-    finally:
-        if sys.stdout.isatty():
-            sys.stdout.write("\033[?25l")
-            sys.stdout.flush()
-    value = value.strip()
-    return value or default
+    finally:
+        if sys.stdout.isatty():
+            sys.stdout.write("\033[?25l")
+            sys.stdout.flush()
+        if raw_tty:
+            raw_tty()
+    value = value.strip()
+    return value or default
 
 
 def portable_provider_menu() -> int:
@@ -8168,22 +8266,23 @@ def portable_prelaunch_menu() -> int:
     enable_ansi()
     main_idx = 7 if settings_ready_except_api_key() else 0
     panel: str | None = None
-    panel_idx = 0
-    panel_rows: list[str] = []
-    panel_values: list[str] = []
+    panel_idx = 0
+    panel_rows: list[str] = []
+    panel_values: list[str] = []
+    panel_last_idx: dict[str, int] = {}
     checks = preflight_lines()
     messages: list[str] = []
     first_render = True
 
-    def open_panel(name: str) -> None:
-        nonlocal panel, panel_idx, panel_rows, panel_values, messages, first_render
-        cfg = load_config()
-        provider, pcfg = get_current_provider(cfg)
-        panel = name
-        panel_idx = 0
-        if name == "language":
-            panel_rows, panel_values = language_panel_rows(cfg)
-            panel_idx = panel_values.index(cfg.get("language", "en"))
+    def open_panel(name: str) -> None:
+        nonlocal panel, panel_idx, panel_rows, panel_values, messages, first_render
+        cfg = load_config()
+        provider, pcfg = get_current_provider(cfg)
+        panel = name
+        panel_idx = panel_last_idx.get(name, 0)
+        if name == "language":
+            panel_rows, panel_values = language_panel_rows(cfg)
+            panel_idx = panel_values.index(cfg.get("language", "en"))
         elif name == "provider":
             panel_rows, panel_values = provider_panel_rows(cfg)
             panel_idx = panel_values.index(provider)
@@ -8205,15 +8304,19 @@ def portable_prelaunch_menu() -> int:
                 panel_rows, panel_values = [f"Advisor model list failed: {type(exc).__name__}: {exc}", "+ Custom advisor model id..."], []
         elif name == "test":
             panel_rows, panel_values = ["Run compatibility test", "Back"], ["run", "back"]
-        elif name == "options":
-            panel_rows, panel_values = llm_option_panel_rows(provider, pcfg, cfg.get("language", "en"))
-        elif name == "preset":
-            panel_rows, panel_values = llm_preset_panel_rows(provider, pcfg, cfg.get("language", "en"))
-
-    def close_panel(next_idx: int | None = None) -> None:
-        nonlocal panel, panel_idx, panel_rows, panel_values, main_idx
-        panel = None
-        panel_idx = 0
+        elif name == "options":
+            panel_rows, panel_values = llm_option_panel_rows(provider, pcfg, cfg.get("language", "en"))
+        elif name == "preset":
+            panel_rows, panel_values = llm_preset_panel_rows(provider, pcfg, cfg.get("language", "en"))
+        if panel_rows:
+            panel_idx = max(0, min(panel_idx, len(panel_rows) - 1))
+
+    def close_panel(next_idx: int | None = None) -> None:
+        nonlocal panel, panel_idx, panel_rows, panel_values, main_idx
+        if panel:
+            panel_last_idx[panel] = panel_idx
+        panel = None
+        panel_idx = 0
         panel_rows = []
         panel_values = []
         if next_idx is not None:
@@ -8236,20 +8339,43 @@ def portable_prelaunch_menu() -> int:
             termios.tcsetattr(fd, termios.TCSANOW, new)
         except Exception:
             fd = -1
-    if sys.stdout.isatty():
-        sys.stdout.write("\033[?25l")
-        sys.stdout.flush()
-    try:
+    if sys.stdout.isatty():
+        sys.stdout.write("\033[?25l")
+        sys.stdout.flush()
+    def restore_line_mode() -> None:
+        if old_settings is not None and fd >= 0:
+            try:
+                import termios
+                termios.tcsetattr(fd, termios.TCSANOW, old_settings)
+            except Exception:
+                pass
+
+    def restore_raw_mode() -> None:
+        if old_settings is not None and fd >= 0:
+            try:
+                import termios
+                new = termios.tcgetattr(fd)
+                new[3] = new[3] & ~(termios.ECHO | termios.ICANON)
+                new[6][termios.VMIN] = 1
+                new[6][termios.VTIME] = 0
+                termios.tcsetattr(fd, termios.TCSANOW, new)
+            except Exception:
+                pass
+
+    try:
         while True:
             first_render = render_prelaunch_screen(main_idx, panel, panel_idx, panel_rows, checks, messages, first_render)
             key = read_menu_key(fd) if fd >= 0 else read_menu_key()
-            if panel:
-                if key in ("up", "k"):
-                    panel_idx = (panel_idx - 1) % max(1, len(panel_rows))
-                    continue
-                if key in ("down", "j"):
-                    panel_idx = (panel_idx + 1) % max(1, len(panel_rows))
-                    continue
+            if panel:
+                panel_name = panel
+                if key in ("up", "k"):
+                    panel_idx = (panel_idx - 1) % max(1, len(panel_rows))
+                    panel_last_idx[panel_name] = panel_idx
+                    continue
+                if key in ("down", "j"):
+                    panel_idx = (panel_idx + 1) % max(1, len(panel_rows))
+                    panel_last_idx[panel_name] = panel_idx
+                    continue
                 if key in ("esc", "left", "q"):
                     close_panel()
                     continue
@@ -8274,7 +8400,7 @@ def portable_prelaunch_menu() -> int:
                         close_panel()
                         continue
                     if value == "__custom__" or panel_idx >= len(panel_values):
-                        model_value = prompt_menu_value("Model id or alias")
+                        model_value = prompt_menu_value("Model id or alias", restore_tty=restore_line_mode, raw_tty=restore_raw_mode)
                     else:
                         model_value = value
                     if model_value:
@@ -8286,7 +8412,7 @@ def portable_prelaunch_menu() -> int:
                         close_panel()
                         continue
                     if value == "__custom__" or panel_idx >= len(panel_values):
-                        advisor_value = prompt_menu_value("Advisor model id", "deepseek-v4-pro")
+                        advisor_value = prompt_menu_value("Advisor model id", "deepseek-v4-pro", restore_tty=restore_line_mode, raw_tty=restore_raw_mode)
                     else:
                         advisor_value = value
                     messages = set_advisor_model_config(advisor_value)
@@ -8296,7 +8422,7 @@ def portable_prelaunch_menu() -> int:
                     if value == "back":
                         close_panel()
                     elif value == "input":
-                        key_value = prompt_menu_value(f"API key for {provider}", secret=True)
+                        key_value = prompt_menu_value(f"API key for {provider}", secret=True, restore_tty=restore_line_mode, raw_tty=restore_raw_mode)
                         if key_value:
                             messages = store_api_key_config(provider, key_value)
                             refresh_checks()
@@ -8307,7 +8433,7 @@ def portable_prelaunch_menu() -> int:
                             "nvidia-hosted": "NVIDIA_API_KEY",
                             "ollama-cloud": "OLLAMA_API_KEY",
                         }.get(provider, "API_KEY")
-                        env_name = prompt_menu_value("Environment variable name", default_env)
+                        env_name = prompt_menu_value("Environment variable name", default_env, restore_tty=restore_line_mode, raw_tty=restore_raw_mode)
                         key_value = os.environ.get(env_name, "").strip()
                         if key_value:
                             messages = store_api_key_config(provider, key_value)
@@ -8320,7 +8446,7 @@ def portable_prelaunch_menu() -> int:
                         if not key_value:
                             messages = ["Clipboard did not contain readable text."]
                         else:
-                            confirm = prompt_menu_value(f"Clipboard contains {mask_secret(key_value)}. Store it? y/N")
+                            confirm = prompt_menu_value(f"Clipboard contains {mask_secret(key_value)}. Store it? y/N", restore_tty=restore_line_mode, raw_tty=restore_raw_mode)
                             if confirm.lower().startswith("y"):
                                 messages = store_api_key_config(provider, key_value)
                             else:
@@ -8336,7 +8462,7 @@ def portable_prelaunch_menu() -> int:
                         close_panel(4)
                     elif value == "edit":
                         default = pcfg.get("base_url") or default_base_url(provider)
-                        url = prompt_menu_value(f"Base URL for {provider}", default)
+                        url = prompt_menu_value(f"Base URL for {provider}", default, restore_tty=restore_line_mode, raw_tty=restore_raw_mode)
                         if url:
                             messages = set_base_url_config(provider, url)
                             refresh_checks()
@@ -8365,21 +8491,27 @@ def portable_prelaunch_menu() -> int:
                             messages = set_llm_option_config(provider, value, "false" if current else "true")
                         except Exception as exc:
                             messages = [f"Option update failed: {type(exc).__name__}: {exc}"]
-                        refresh_checks()
-                        cfg = load_config()
-                        provider, pcfg = get_current_provider(cfg)
-                        panel_rows, panel_values = llm_option_panel_rows(provider, pcfg, cfg.get("language", "en"))
-                    else:
-                        default = llm_option_prompt_default(provider, pcfg, value)
-                        entered = prompt_menu_value(f"{value} for {provider} (default/unset clears)", default)
+                        refresh_checks()
+                        cfg = load_config()
+                        provider, pcfg = get_current_provider(cfg)
+                        old_idx = panel_idx
+                        panel_rows, panel_values = llm_option_panel_rows(provider, pcfg, cfg.get("language", "en"))
+                        panel_idx = max(0, min(old_idx, len(panel_rows) - 1))
+                        panel_last_idx["options"] = panel_idx
+                    else:
+                        default = llm_option_prompt_default(provider, pcfg, value)
+                        entered = prompt_menu_value(f"{value} for {provider} (default/unset clears)", default, restore_tty=restore_line_mode, raw_tty=restore_raw_mode)
                         try:
                             messages = set_llm_option_config(provider, value, entered)
                         except Exception as exc:
                             messages = [f"Option update failed: {type(exc).__name__}: {exc}"]
-                        refresh_checks()
-                        cfg = load_config()
-                        provider, pcfg = get_current_provider(cfg)
-                        panel_rows, panel_values = llm_option_panel_rows(provider, pcfg, cfg.get("language", "en"))
+                        refresh_checks()
+                        cfg = load_config()
+                        provider, pcfg = get_current_provider(cfg)
+                        old_idx = panel_idx
+                        panel_rows, panel_values = llm_option_panel_rows(provider, pcfg, cfg.get("language", "en"))
+                        panel_idx = max(0, min(old_idx, len(panel_rows) - 1))
+                        panel_last_idx["options"] = panel_idx
                 elif panel == "preset":
                     if value == "back":
                         open_panel("options")
@@ -8391,12 +8523,14 @@ def portable_prelaunch_menu() -> int:
                         except Exception as exc:
                             messages = [f"Preset failed: {type(exc).__name__}: {exc}"]
                         refresh_checks()
-                        cfg = load_config()
-                        provider, pcfg = get_current_provider(cfg)
-                        panel = "options"
-                        panel_idx = 0
-                        panel_rows, panel_values = llm_option_panel_rows(provider, pcfg, cfg.get("language", "en"))
-                continue
+                        cfg = load_config()
+                        provider, pcfg = get_current_provider(cfg)
+                        panel = "options"
+                        panel_idx = panel_last_idx.get("options", 0)
+                        panel_rows, panel_values = llm_option_panel_rows(provider, pcfg, cfg.get("language", "en"))
+                        panel_idx = max(0, min(panel_idx, len(panel_rows) - 1))
+                        panel_last_idx["options"] = panel_idx
+                continue
 
             if key in ("up", "k"):
                 main_idx = (main_idx - 1) % 10

package/docs/README.ja.md

@@ -47,7 +47,7 @@ vLLM、NVIDIA hosted、self-hosted NIM を選択し、通常の Claude Code 引
 
 Credits: One Ciel LLC
 
-現在のバージョン: `0.1.38`
+現在のバージョン: `0.1.42`
 
 ## 作られた理由
 
@@ -351,6 +351,30 @@ Windows/Linux 管理、クリーンアップスクリプト、定期的なセキ
 
 ## 変更履歴
 
+### 0.1.42
+
+- **ライブストリーム進捗**: statusline が upstream streaming の出力進捗を
+  入力/出力 token 推定値と chunk 数で継続更新します。
+
+### 0.1.41
+
+- **Statusline 表示改善**: upstream token 数に桁区切りと `tok` 前の空白を入れ、
+  `27,501 tok` のように表示します。
+
+### 0.1.40
+
+- **RPM 0 を保持**: `rate_limit_rpm=0` の設定が provider 既定値に戻らず、
+  明示的な無制限モードとして保存されます。
+
+### 0.1.39
+
+- **メニュー入力修正**: テキスト/数字プロンプトの前に terminal line/echo mode を
+  復元し、prelaunch UI で入力した数字が見えるようにしました。
+- **数字検証の安定化**: 数字オプションへ不正な文字を入れてもメニューが
+  クラッシュせず、案内メッセージを表示します。
+- **プリセット表示改善**: preset 適用後に実際の context、reserve、output、
+  timeout 値をメッセージに表示します。
+
 ### 0.1.38
 
 - **ユーザー選択の context window を優先**: NVIDIA hosted の 32K safety cap を

package/docs/README.ko.md

@@ -47,7 +47,7 @@ NVIDIA hosted, self-hosted NIM을 선택하고, Claude Code의 일반 인자는
 
 Credits: One Ciel LLC
 
-현재 버전: `0.1.38`
+현재 버전: `0.1.42`
 
 ## 왜 만들었나
 
@@ -351,6 +351,30 @@ Windows 이벤트 로그 리뷰, 바이러스/랜섬웨어 침입 시도 정리,
 
 ## 변경 이력
 
+### 0.1.42
+
+- **실시간 스트림 진행 표시**: statusline이 upstream streaming 출력 진행을
+  입력/출력 token 추정치와 chunk 수로 계속 갱신합니다.
+
+### 0.1.41
+
+- **Statusline 표시 개선**: upstream token 수에 천 단위 구분자와 `tok` 앞 공백을
+  넣어 `27,501 tok`처럼 표시합니다.
+
+### 0.1.40
+
+- **RPM 0 유지**: `rate_limit_rpm=0` 설정이 provider 기본값으로 되돌아가지 않고
+  명시적인 무제한 모드로 저장됩니다.
+
+### 0.1.39
+
+- **메뉴 입력 수정**: 텍스트/숫자 프롬프트 전에 터미널 line/echo 모드를 복구하여
+  prelaunch UI에서 입력한 숫자가 보이게 했습니다.
+- **숫자 검증 안정화**: 숫자 옵션에 잘못된 문자를 넣어도 메뉴가 크래시되지 않고
+  안내 메시지를 표시합니다.
+- **프리셋 표시 개선**: preset 적용 후 실제 context, reserve, output, timeout 값을
+  메시지에 표시합니다.
+
 ### 0.1.38
 
 - **사용자 선택 context window 우선**: NVIDIA hosted 32K safety cap을 제거했습니다.

package/docs/README.zh.md

@@ -47,7 +47,7 @@ NIM，并把普通 Claude Code 参数原样传递。
 
 Credits: One Ciel LLC
 
-当前版本: `0.1.38`
+当前版本: `0.1.42`
 
 ## 为什么存在
 
@@ -337,6 +337,29 @@ Hermes 格式模型或部分较旧的 Qwen tool template。
 
 ## 更新日志
 
+### 0.1.42
+
+- **实时流式进度**：statusline 会持续更新 upstream streaming 输出进度，
+  显示输入/输出 token 估算值和 chunk 数。
+
+### 0.1.41
+
+- **Statusline 格式优化**：upstream token 数现在带千位分隔符，并在 `tok` 前加入空格，
+  例如 `27,501 tok`。
+
+### 0.1.40
+
+- **保留 RPM 0**：`rate_limit_rpm=0` 现在会保存为明确的无限制模式，
+  不会回退到 provider 默认值。
+
+### 0.1.39
+
+- **菜单输入修复**：在文本/数字提示前恢复 terminal line/echo mode，
+  prelaunch UI 中输入的数字现在可见。
+- **数字校验更稳**：数字选项输入非法字符时不再让菜单崩溃，而是显示提示消息。
+- **Preset 可见性**：应用 preset 后会显示实际 context、reserve、output、
+  timeout 值。
+
 ### 0.1.38
 
 - **优先使用用户选择的 context window**：移除 NVIDIA hosted 的 32K safety cap。

package/docs/manual.md

@@ -10,7 +10,7 @@ Code starts, while passing normal Claude Code arguments through unchanged.
 
 Credits: One Ciel LLC
 
-Current version: `0.1.38`
+Current version: `0.1.42`
 
 ## Install
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@oneciel-ai/claude-any",
-  "version": "0.1.38",
+  "version": "0.1.42",
   "description": "Claude Code provider selector for Anthropic, Ollama, Ollama Cloud, vLLM, NVIDIA hosted, and self-hosted NIM.",
   "license": "MIT",
   "author": "One Ciel LLC",