spawnllm 0.13.4__tar.gz → 0.15.0__tar.gz

{spawnllm-0.13.4 → spawnllm-0.15.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: spawnllm
-Version: 0.13.4
+Version: 0.15.0
 Summary: Delete your subprocess wrappers around claude, codex, and gemini.
 Keywords: 
 Author: Yasyf Mohamedali

{spawnllm-0.13.4 → spawnllm-0.15.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "spawnllm"
-version = "0.13.4"
+version = "0.15.0"
 description = "Delete your subprocess wrappers around claude, codex, and gemini."
 readme = "README.md"
 license = "MIT"

{spawnllm-0.13.4 → spawnllm-0.15.0}/pyproject.toml.orig

@@ -1,7 +1,7 @@
 [project]
 name = "spawnllm"
 # Inert sentinel: the real version is set from the release tag (uv version --frozen).
-version = "0.13.4"
+version = "0.15.0"
 description = "Delete your subprocess wrappers around claude, codex, and gemini."
 readme = "README.md"
 license = "MIT"

{spawnllm-0.13.4 → spawnllm-0.15.0}/spawnllm/__init__.py

@@ -34,7 +34,7 @@ from spawnllm.extract import extract, extract_sync
 from spawnllm.response import DiscardedAttempt, Error, Output, Response, Result
 from spawnllm.run import run, run_sync
 from spawnllm.spec import AppleConfig, ClaudeConfig, CodexConfig, GeminiConfig, RunSpec
-from spawnllm.types import ProviderName, TModel, TSpecialty
+from spawnllm.types import ProviderName, TModel, TReasoningEffort, TSpecialty
 
 __all__ = [
     "AntigravityCliBackend",
@@ -66,6 +66,7 @@ __all__ = [
     "Result",
     "RunSpec",
     "TModel",
+    "TReasoningEffort",
     "TSpecialty",
     "call",
     "call_sync",

{spawnllm-0.13.4 → spawnllm-0.15.0}/spawnllm/backends/codex.py

@@ -29,8 +29,8 @@ class CodexCliBackend(CliBackend):
     """
 
     models: ClassVar[dict[TModel, str]] = {
-        "small": "gpt-5.6-luna:low",
-        "medium": "gpt-5.6-luna:medium",
+        "small": "gpt-6-luna:low",
+        "medium": "gpt-6-luna:medium",
         "large": "gpt-5.5:medium",
     }
     provider: ClassVar[ProviderName] = "codex"

{spawnllm-0.13.4 → spawnllm-0.15.0}/spawnllm/backends/openai_endpoint.py

@@ -12,7 +12,7 @@ if TYPE_CHECKING:
     from spawnllm.backends.base import BackendStatus
     from spawnllm.response import Response
     from spawnllm.spec import RunSpec
-    from spawnllm.types import ProviderName, TModel
+    from spawnllm.types import ProviderName, TModel, TReasoningEffort
 
 
 class OpenAiEndpointBackend(LlmBackend):
@@ -30,6 +30,9 @@ class OpenAiEndpointBackend(LlmBackend):
         model: The literal model id sent in every request body.
         api_key: Bearer token for the `Authorization` header; defaults to
             `"local"` for self-hosted servers that ignore it.
+        reasoning_effort: The `reasoning_effort` sent in every request body;
+            `None` omits it and leaves the server's default, which on a
+            reasoning model can spend most of a call's latency thinking.
         transport: Async transport injected into the `httpx.AsyncClient` used by
             `aexecute` — e.g. a record/replay caching transport; `None` uses
             httpx's default transport. The synchronous `execute` path always uses
@@ -44,17 +47,29 @@ class OpenAiEndpointBackend(LlmBackend):
     schema_dialect: ClassVar[str | None] = "openai"
 
     def __init__(
-        self, base_url: str, model: str, *, api_key: str = "local", transport: httpx.AsyncBaseTransport | None = None
+        self,
+        base_url: str,
+        model: str,
+        *,
+        api_key: str = "local",
+        reasoning_effort: TReasoningEffort | None = None,
+        transport: httpx.AsyncBaseTransport | None = None,
     ) -> None:
         self.base_url = base_url.rstrip("/")
         self.model = model
         self.api_key = api_key
+        self.reasoning_effort = reasoning_effort
         self.transport = transport
         self.models: dict[TModel, str] = {"small": model, "medium": model, "large": model}
 
     def openai_section(self) -> dict[str, Any]:
         """Return the `openai_endpoint` wire section the core turns into the HTTP request."""
-        return {"api_key": self.api_key, "base_url": self.base_url, "model": self.model}
+        return {
+            "api_key": self.api_key,
+            "base_url": self.base_url,
+            "model": self.model,
+            "reasoning_effort": self.reasoning_effort,
+        }
 
     def resolve(self, resp: httpx.Response, spec: RunSpec) -> Response:
         return self.to_response(

{spawnllm-0.13.4 → spawnllm-0.15.0}/spawnllm/types.py

@@ -4,7 +4,7 @@ from __future__ import annotations
 
 from typing import Literal
 
-__all__ = ["ProviderName", "TModel", "TSettingSource", "TSpecialty"]
+__all__ = ["ProviderName", "TModel", "TReasoningEffort", "TSettingSource", "TSpecialty"]
 
 TSpecialty = Literal["debugging", "review", "general"]
 """Task specialty; `LlmBackends.for_specialty` maps each to its registered backend."""
@@ -15,5 +15,8 @@ TSettingSource = Literal["user", "project", "local"]
 TModel = Literal["small", "medium", "large"]
 """Abstract model tier; each backend maps it to a provider-specific model name."""
 
+TReasoningEffort = Literal["none", "minimal", "low", "medium", "high", "xhigh"]
+"""A `reasoning_effort` an OpenAI-compatible endpoint accepts; `OpenAiEndpointBackend` sends it with every request."""
+
 ProviderName = Literal["claude", "claude-sdk", "codex", "gemini", "antigravity", "mlx", "apple", "openai_endpoint"]
 """Backend provider identifier; keys the per-backend `provider_configs` on a `RunSpec`."""