Skip to content

Hypergraph is more powerful. Transform unstructured text into structured knowledge with LLMs. Graphs, hypergraphs, and spatio-temporal extractions — with one command.

fix: resolve provider API key env vars in create_llm OpenAI-compatible path

Frankie-Xu6ff362287734
diff --git a/hyperextract/utils/client.py b/hyperextract/utils/client.py
index dc0354c..b270f26 100644
--- a/hyperextract/utils/client.py
+++ b/hyperextract/utils/client.py
@@ -87,10 +87,12 @@ def _env_api_key(provider: str) -> str:
Falls back to OPENAI_API_KEY for OpenAI-compatible providers.
"""
- for var in PROVIDER_API_KEY_ENV.get(provider, ("OPENAI_API_KEY",)):
+ for var in PROVIDER_API_KEY_ENV.get(provider, ()):
value = os.environ.get(var, "")
if value:
return value
+ if provider not in ANTHROPIC_PROVIDERS:
+ return os.environ.get("OPENAI_API_KEY", "")
return ""
@@ -379,7 +381,7 @@ def create_llm(
chat_kwargs: dict[str, Any] = {
"model": config["model"],
- "api_key": config["api_key"] or os.environ.get("OPENAI_API_KEY", ""),
+ "api_key": config["api_key"] or _env_api_key(provider),
"base_url": config.get("base_url") or None,
"temperature": config.get("temperature", 0),
}
@@ -421,10 +423,12 @@ def create_embedder(
base_url = config.get("base_url", "")
uses_custom = bool(base_url and base_url.rstrip("/") != OPENAI_API_URL)
+ resolved_key = config["api_key"] or _env_api_key(config.get("provider", ""))
+
if uses_custom:
return CompatibleEmbeddings(
model=config["model"],
- api_key=config["api_key"] or os.environ.get("OPENAI_API_KEY", ""),
+ api_key=resolved_key,
base_url=base_url,
**kwargs,
)
@@ -433,7 +437,7 @@ def create_embedder(
return OpenAIEmbeddings(
model=config["model"],
- api_key=config["api_key"] or os.environ.get("OPENAI_API_KEY", ""),
+ api_key=resolved_key,
**kwargs,
)