@oneciel-ai/claude-any 0.1.36 → 0.1.38

package/README.md

@@ -48,7 +48,7 @@ arguments through unchanged.
 
 Credits: One Ciel LLC
 
-Current version: `0.1.36`
+Current version: `0.1.38`
 
 ## Why This Exists
 
@@ -381,6 +381,22 @@ steps under that larger model's supervision.
 
 ## Changelog
 
+### 0.1.38
+
+- **User-selected context windows**: removes the NVIDIA hosted 32K safety cap.
+  The router now uses the context window selected in LLM options or headless
+  configuration, with model-aware fallback only when no value is configured.
+- **NVIDIA presets updated**: NVIDIA hosted presets now start at 65K and scale
+  up to 256K for large-output/reasoning workflows.
+
+### 0.1.37
+
+- **Pseudo tool-call recovery**: the NVIDIA/OpenAI-compatible stream path now
+  suppresses `<|tool_calls_section_begin|>...` pseudo tool-call text and
+  converts it back into Claude `tool_use` blocks when possible.
+- **Streaming defaults**: provider streaming defaults to on; NVIDIA hosted
+  remains forced to the streaming upstream path for stability.
+
 ### 0.1.36
 
 - **NVIDIA upstream streaming**: NVIDIA hosted router calls now use upstream

package/claude_any.py

@@ -85,7 +85,7 @@ PROVIDER_LABELS = {
     "self-hosted-nim": "Self Hosted NIM",
 }
 APP_NAME = "Claude Any"
-VERSION = "0.1.36"
+VERSION = "0.1.38"
 CREDITS = "Credits: One Ciel LLC"
 
 LOG_LEVELS = {"SILENT": 0, "ERROR": 1, "WARN": 2, "INFO": 3, "DEBUG": 4, "TRACE": 5}
@@ -144,7 +144,7 @@ LANGUAGES = {
     "zh": "中文",
 }
 
-MODEL_PRESETS: dict[str, dict[str, Any]] = {
+MODEL_PRESETS: dict[str, dict[str, Any]] = {
     "glm-4.7": {"compat_max_tokens": 64, "thinking": True, "num_ctx_min": 32768, "num_ctx_max": 131072},
     "glm-5.1": {"compat_max_tokens": 64, "thinking": True, "num_ctx_min": 32768, "num_ctx_max": 131072},
     "glm-4.7:cloud": {"compat_max_tokens": 64, "thinking": True, "num_ctx_min": 32768, "num_ctx_max": 131072},
@@ -154,10 +154,21 @@ MODEL_PRESETS: dict[str, dict[str, Any]] = {
     "qwen3.6:27b": {"compat_max_tokens": 16, "thinking": False, "num_ctx_min": 32768, "num_ctx_max": 65536},
     "deepseek-r1": {"compat_max_tokens": 64, "thinking": True, "num_ctx_min": 32768, "num_ctx_max": 131072},
     "llama3.3:70b": {"compat_max_tokens": 16, "thinking": False, "num_ctx_min": 32768, "num_ctx_max": 131072},
-}
-
-
-def model_preset(model_id: str) -> dict[str, Any]:
+}
+
+
+def nvidia_hosted_context_default(model_id: str) -> int:
+    model = model_id.lower()
+    if "kimi-k2.6" in model or "kimi_k2.6" in model:
+        return 262144
+    if "deepseek" in model:
+        return 131072
+    if "glm" in model or "qwen" in model:
+        return 65536
+    return 65536
+
+
+def model_preset(model_id: str) -> dict[str, Any]:
     """Return preset dict for a model ID, checking exact match then prefix match."""
     if model_id in MODEL_PRESETS:
         return MODEL_PRESETS[model_id]
@@ -722,7 +733,7 @@ DEFAULT_CONFIG: dict[str, Any] = {
             "native_compat": False,
             "rate_limit_rpm": 40,
             "rate_limit_status": True,
-            "context_window": 32768,
+            "context_window": 65536,
             "max_output_tokens": 4096,
             "temperature": 0.7,
             "top_p": 0.8,
@@ -788,7 +799,21 @@ def apply_config_migrations(cfg: dict[str, Any]) -> None:
     if not migrations.get(marker):
         pcfg = cfg.get("providers", {}).get("nvidia-hosted", {})
         if isinstance(pcfg, dict) and not positive_int(pcfg.get("context_window")):
-            pcfg["context_window"] = 32768
+            pcfg["context_window"] = nvidia_hosted_context_default(str(pcfg.get("current_model") or ""))
+        migrations[marker] = True
+
+    marker = "nvidia_context_window_unforce_32k_20260513"
+    if not migrations.get(marker):
+        pcfg = cfg.get("providers", {}).get("nvidia-hosted", {})
+        if isinstance(pcfg, dict) and positive_int(pcfg.get("context_window")) == 32768:
+            pcfg["context_window"] = nvidia_hosted_context_default(str(pcfg.get("current_model") or ""))
+        migrations[marker] = True
+
+    marker = "stream_enabled_default_true_20260513"
+    if not migrations.get(marker):
+        for pcfg in (cfg.get("providers") or {}).values():
+            if isinstance(pcfg, dict) and "stream_enabled" not in pcfg:
+                pcfg["stream_enabled"] = True
         migrations[marker] = True
 
 
@@ -3613,7 +3638,7 @@ def openai_context_limit_for_budget(provider: str, pcfg: dict[str, Any]) -> int:
     if configured:
         return configured
     if provider == "nvidia-hosted":
-        return 32768
+        return nvidia_hosted_context_default(str(pcfg.get("current_model") or ""))
     return 65536
 
 
@@ -3988,7 +4013,7 @@ def maybe_handle_advisor_request(handler: BaseHTTPRequestHandler, provider: str,
     return True
 
 
-def normalize_tool_arguments(tool_name: str, args: Any) -> dict[str, Any]:
+def normalize_tool_arguments(tool_name: str, args: Any) -> dict[str, Any]:
     if isinstance(args, dict):
         return args
     if isinstance(args, str):
@@ -4003,17 +4028,86 @@ def normalize_tool_arguments(tool_name: str, args: Any) -> dict[str, Any]:
             pass
         if tool_name == "Bash":
             return {"command": text}
-    return {}
-
-
-def ollama_chat_to_anthropic(data: dict[str, Any], model: str, source_body: dict[str, Any] | None = None) -> dict[str, Any]:
-    message = data.get("message") if isinstance(data.get("message"), dict) else {}
-    content: list[dict[str, Any]] = []
-    text = message.get("content") or ""
-    if text:
-        content.append({"type": "text", "text": text})
-    tool_id_prefix = f"toolu_ollama_{int(time.time() * 1000)}_{os.getpid()}"
-    for i, call in enumerate(message.get("tool_calls") or []):
+    return {}
+
+
+PSEUDO_TOOL_START = "<|tool_calls_section_begin|>"
+PSEUDO_TOOL_END = "<|tool_calls_section_end|>"
+PSEUDO_CALL_BEGIN = "<|tool_call_begin|>"
+PSEUDO_ARG_BEGIN = "<|tool_call_argument_begin|>"
+PSEUDO_CALL_END = "<|tool_call_end|>"
+
+
+def infer_tool_name_from_args(args: dict[str, Any]) -> str:
+    keys = set(args)
+    if "command" in keys:
+        return "Bash"
+    if {"file_path", "content"}.issubset(keys):
+        return "Write"
+    if {"file_path", "old_string", "new_string"}.issubset(keys):
+        return "Edit"
+    if "file_path" in keys:
+        return "Read"
+    if "taskId" in keys and "status" in keys:
+        return "TaskUpdate"
+    return "TaskList" if not args else "Write"
+
+
+def parse_pseudo_tool_calls(text: str) -> tuple[str, list[dict[str, Any]]]:
+    if PSEUDO_TOOL_START not in text:
+        return text, []
+    visible_parts: list[str] = []
+    calls: list[dict[str, Any]] = []
+    pos = 0
+    while True:
+        start = text.find(PSEUDO_TOOL_START, pos)
+        if start < 0:
+            visible_parts.append(text[pos:])
+            break
+        visible_parts.append(text[pos:start])
+        end = text.find(PSEUDO_TOOL_END, start)
+        if end < 0:
+            section = text[start + len(PSEUDO_TOOL_START):]
+            pos = len(text)
+        else:
+            section = text[start + len(PSEUDO_TOOL_START):end]
+            pos = end + len(PSEUDO_TOOL_END)
+        for match in re.finditer(
+            re.escape(PSEUDO_CALL_BEGIN) + r"(.*?)" + re.escape(PSEUDO_ARG_BEGIN) + r"(.*?)" + re.escape(PSEUDO_CALL_END),
+            section,
+            flags=re.DOTALL,
+        ):
+            raw_header = match.group(1).strip()
+            raw_args = match.group(2).strip()
+            try:
+                args = json.loads(raw_args)
+            except Exception:
+                continue
+            if not isinstance(args, dict):
+                continue
+            name = ""
+            for part in re.split(r"[\s:|,]+", raw_header):
+                candidate = _fuzzy_match_tool_name(part)
+                if candidate:
+                    name = candidate
+                    break
+            if not name:
+                name = infer_tool_name_from_args(args)
+            calls.append({"function": {"name": name, "arguments": args}, "id": raw_header})
+        if end < 0:
+            break
+    return "".join(visible_parts), calls
+
+
+def ollama_chat_to_anthropic(data: dict[str, Any], model: str, source_body: dict[str, Any] | None = None) -> dict[str, Any]:
+    message = data.get("message") if isinstance(data.get("message"), dict) else {}
+    content: list[dict[str, Any]] = []
+    text = message.get("content") or ""
+    text, pseudo_tool_calls = parse_pseudo_tool_calls(text)
+    if text:
+        content.append({"type": "text", "text": text})
+    tool_id_prefix = f"toolu_ollama_{int(time.time() * 1000)}_{os.getpid()}"
+    for i, call in enumerate(list(message.get("tool_calls") or []) + pseudo_tool_calls):
         fn = call.get("function") if isinstance(call, dict) else {}
         if not isinstance(fn, dict) or not fn.get("name"):
             continue
@@ -4609,6 +4703,8 @@ def stream_openai_chat_to_anthropic_sse(
     text_suppressed_for_plan = False
     text_index: int | None = None
     text_so_far = ""
+    pseudo_text = ""
+    pseudo_mode = False
     text_buffer = ""
     tool_fragments: dict[int, dict[str, Any]] = {}
     output_tokens = 0
@@ -4667,6 +4763,21 @@ def stream_openai_chat_to_anthropic_sse(
             delta = choice.get("delta") if isinstance(choice.get("delta"), dict) else {}
             text_chunk = delta.get("content") or ""
             if text_chunk:
+                if pseudo_mode or PSEUDO_TOOL_START in text_chunk:
+                    before, sep, after = text_chunk.partition(PSEUDO_TOOL_START)
+                    if before and not pseudo_mode:
+                        text_so_far += before
+                        if word_chunking:
+                            text_buffer += before
+                            to_flush, text_buffer = _split_word_buffer(text_buffer, force=False)
+                            emit_text_delta(to_flush)
+                        else:
+                            emit_text_delta(before)
+                    pseudo_mode = True
+                    pseudo_text += (sep + after) if sep else text_chunk
+                    if PSEUDO_TOOL_END in pseudo_text:
+                        pseudo_mode = False
+                    continue
                 if source_body is not None and not text_started and not tool_fragments and should_auto_enter_plan_mode(source_body, text_so_far + text_chunk, []):
                     text_so_far += text_chunk
                     text_suppressed_for_plan = True
@@ -4698,6 +4809,15 @@ def stream_openai_chat_to_anthropic_sse(
             emit_text_delta(to_flush)
 
         tool_calls: list[dict[str, Any]] = []
+        _, pseudo_tool_calls = parse_pseudo_tool_calls(pseudo_text)
+        for i, pseudo in enumerate(pseudo_tool_calls):
+            fn = pseudo.get("function") if isinstance(pseudo, dict) else {}
+            if isinstance(fn, dict):
+                tool_fragments.setdefault(100000 + i, {
+                    "id": str(pseudo.get("id") or ""),
+                    "name": str(fn.get("name") or ""),
+                    "arguments": json.dumps(fn.get("arguments") or {}, ensure_ascii=False),
+                })
         for _, fragment in sorted(tool_fragments.items()):
             raw_name = str(fragment.get("name") or "")
             if not raw_name:
@@ -4964,7 +5084,8 @@ def forward_openai_compatible_chat(handler: BaseHTTPRequestHandler, provider: st
         model = ncp_model_id_for_nvidia_hosted(model)
     url = join_url(provider_upstream_request_base(provider, pcfg), "/chat/completions")
     waited, rpm_used, rpm_limit = apply_router_rate_limit(provider, pcfg, model)
-    stream = bool(body.get("stream", True))
+    stream_enabled = bool(pcfg.get("stream_enabled", True))
+    stream = True if provider == "nvidia-hosted" else bool(body.get("stream", stream_enabled)) and stream_enabled
     notice = rate_limit_notice(waited, rpm_used, rpm_limit, bool(pcfg.get("rate_limit_status", True)))
     if stream:
         req_body = openai_compatible_chat_request(provider, model, body, pcfg, stream=True)
@@ -6021,14 +6142,72 @@ def apply_llm_preset_to_provider(provider: str, pcfg: dict[str, Any], preset_id:
         }
         for token in tokens_by_preset[preset_id]:
             apply_provider_option(provider, pcfg, token)
-    else:
-        native_default = "false" if provider == "nvidia-hosted" else "true"
-        server_limit = upstream_model_context_limit(provider, pcfg) if provider in ("vllm", "self-hosted-nim") else None
-        tokens_by_preset = {
-            "balanced": [
-                "context_window=32768",
-                "reserve=2048",
-                "max_output_tokens=4096",
+    else:
+        native_default = "false" if provider == "nvidia-hosted" else "true"
+        server_limit = upstream_model_context_limit(provider, pcfg) if provider in ("vllm", "self-hosted-nim") else None
+        if provider == "nvidia-hosted":
+            tokens_by_preset = {
+                "balanced": [
+                    "context_window=65536",
+                    "reserve=4096",
+                    "max_output_tokens=4096",
+                    "timeout=300000",
+                    "temperature=0.3",
+                    "unset:top_p",
+                    "unset:top_k",
+                ],
+                "coding": [
+                    "context_window=65536",
+                    "reserve=4096",
+                    "max_output_tokens=4096",
+                    "timeout=300000",
+                    "temperature=0.2",
+                    "unset:top_p",
+                    "unset:top_k",
+                ],
+                "fast": [
+                    "context_window=65536",
+                    "reserve=2048",
+                    "max_output_tokens=2048",
+                    "timeout=300000",
+                    "temperature=0.2",
+                    "unset:top_p",
+                    "unset:top_k",
+                ],
+                "long-context-65k": [
+                    "context_window=131072",
+                    "reserve=8192",
+                    "max_output_tokens=4096",
+                    "timeout=900000",
+                    "temperature=0.3",
+                    "unset:top_p",
+                    "unset:top_k",
+                ],
+                "large-output": [
+                    "context_window=262144",
+                    "reserve=8192",
+                    "max_output_tokens=8192",
+                    "timeout=1200000",
+                    "temperature=0.3",
+                    "unset:top_p",
+                    "unset:top_k",
+                ],
+                "reasoning": [
+                    "context_window=262144",
+                    "reserve=8192",
+                    "max_output_tokens=4096",
+                    "timeout=1800000",
+                    "temperature=0.6",
+                    "unset:top_p",
+                    "unset:top_k",
+                ],
+            }
+        else:
+            tokens_by_preset = {
+            "balanced": [
+                "context_window=32768",
+                "reserve=2048",
+                "max_output_tokens=4096",
                 "timeout=300000",
                 "temperature=0.3",
                 "unset:top_p",
@@ -6082,10 +6261,10 @@ def apply_llm_preset_to_provider(provider: str, pcfg: dict[str, Any], preset_id:
                 "timeout=1800000",
                 "temperature=0.6",
                 "unset:top_p",
-                "unset:top_k",
-                f"native={native_default}",
-            ],
-        }
+                "unset:top_k",
+                f"native={native_default}",
+            ],
+            }
         for token in tokens_by_preset[preset_id]:
             if provider == "nvidia-hosted" and token.startswith("native="):
                 continue

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: `0.1.36`
+現在のバージョン: `0.1.38`
 
 ## 作られた理由
 
@@ -351,6 +351,22 @@ 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
 
+### 0.1.38
+
+- **ユーザー選択の context window を優先**: NVIDIA hosted の 32K safety cap を
+  削除しました。router は LLM options または headless 設定で選ばれた
+  context window を使い、未設定の場合のみモデル別 fallback を使います。
+- **NVIDIA preset 更新**: NVIDIA hosted preset は 65K から開始し、
+  large-output/reasoning workflow では 256K まで使います。
+
+### 0.1.37
+
+- **Pseudo tool-call recovery**: NVIDIA/OpenAI-compatible stream 経路で
+  `<|tool_calls_section_begin|>...` pseudo tool-call テキストを画面に出さず、
+  可能な場合は Claude `tool_use` ブロックへ復元します。
+- **Streaming defaults**: provider streaming の既定値は on です。NVIDIA hosted
+  は安定性のため upstream streaming 経路に固定されます。
+
 ### 0.1.36
 
 - **NVIDIA upstream streaming**: NVIDIA hosted router 呼び出しは upstream にも

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: `0.1.36`
+현재 버전: `0.1.38`
 
 ## 왜 만들었나
 
@@ -351,6 +351,22 @@ Windows 이벤트 로그 리뷰, 바이러스/랜섬웨어 침입 시도 정리,
 
 ## 변경 이력
 
+### 0.1.38
+
+- **사용자 선택 context window 우선**: NVIDIA hosted 32K safety cap을 제거했습니다.
+  router는 LLM 옵션 또는 headless 설정에서 선택한 context window를 사용하고,
+  값이 없을 때만 모델별 fallback을 사용합니다.
+- **NVIDIA preset 업데이트**: NVIDIA hosted preset은 65K부터 시작하고,
+  large-output/reasoning 워크플로에서는 256K까지 사용합니다.
+
+### 0.1.37
+
+- **Pseudo tool-call recovery**: NVIDIA/OpenAI-compatible stream 경로에서
+  `<|tool_calls_section_begin|>...` pseudo tool-call 텍스트를 화면에 출력하지
+  않고 가능한 경우 Claude `tool_use` 블록으로 복구합니다.
+- **Streaming defaults**: provider streaming 기본값은 on이며, NVIDIA hosted는
+  안정성을 위해 upstream streaming 경로로 고정됩니다.
+
 ### 0.1.36
 
 - **NVIDIA upstream streaming**: NVIDIA hosted router 호출은 이제 upstream에도

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: `0.1.36`
+当前版本: `0.1.38`
 
 ## 为什么存在
 
@@ -337,6 +337,22 @@ Hermes 格式模型或部分较旧的 Qwen tool template。
 
 ## 更新日志
 
+### 0.1.38
+
+- **优先使用用户选择的 context window**：移除 NVIDIA hosted 的 32K safety cap。
+  router 会使用 LLM options 或 headless 配置中选择的 context window，
+  只有未配置时才使用按模型推断的 fallback。
+- **NVIDIA preset 更新**：NVIDIA hosted preset 从 65K 起步，
+  large-output/reasoning 工作流最高使用 256K。
+
+### 0.1.37
+
+- **Pseudo tool-call recovery**：NVIDIA/OpenAI-compatible stream 路径现在会
+  隐藏 `<|tool_calls_section_begin|>...` pseudo tool-call 文本，并尽可能恢复为
+  Claude `tool_use` block。
+- **Streaming defaults**：provider streaming 默认开启；NVIDIA hosted 为了稳定性
+  固定使用 upstream streaming 路径。
+
 ### 0.1.36
 
 - **NVIDIA upstream streaming**：NVIDIA hosted router 调用现在也会向 upstream

package/docs/manual.md

@@ -10,7 +10,7 @@ Code starts, while passing normal Claude Code arguments through unchanged.
 
 Credits: One Ciel LLC
 
-Current version: `0.1.36`
+Current version: `0.1.38`
 
 ## Install
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@oneciel-ai/claude-any",
-  "version": "0.1.36",
+  "version": "0.1.38",
   "description": "Claude Code provider selector for Anthropic, Ollama, Ollama Cloud, vLLM, NVIDIA hosted, and self-hosted NIM.",
   "license": "MIT",
   "author": "One Ciel LLC",