Hypergraph is more powerful. Transform unstructured text into structured knowledge with LLMs. Graphs, hypergraphs, and spatio-temporal extractions — with one command.
docs: add Gemini provider example and README entry
Frankie-Xua5fe746fb8a2
diff --git a/README.md b/README.mdindex 504400d..9bd8987 100644--- a/README.md+++ b/README.md@@ -83,7 +83,7 @@ he config embedder -p vllm -u http://localhost:8001/v1 -k dummy -m BAAI/bge-m3```<details>-<summary><b>More providers</b> — Anthropic (Claude), Alibaba Bailian, OrcaRouter…</summary>+<summary><b>More providers</b> — Anthropic (Claude), Google Gemini, Alibaba Bailian, OrcaRouter…</summary><br>```bash@@ -91,11 +91,15 @@ he config embedder -p vllm -u http://localhost:8001/v1 -k dummy -m BAAI/bge-m3he config llm -p anthropic -k YOUR_ANTHROPIC_API_KEYhe config embedder -p openai -k YOUR_OPENAI_API_KEY+# Google Gemini — LLM only (default: gemini-3.8-flash)+he config llm -p google -k YOUR_GOOGLE_API_KEY+he config embedder -p openai -k YOUR_OPENAI_API_KEY+# Alibaba Bailian (Qwen, LLM + embeddings in one key)he config init -p bailian -k YOUR_BAILIAN_API_KEY```-> OpenAI and Bailian provide both LLM and embedding models in one API. Anthropic and DeepSeek are LLM-only (pair them with an OpenAI-compatible embedder). DeepSeek is the most cost-effective option (~$0.001-0.005/page).+> OpenAI and Bailian provide both LLM and embedding models in one API. Anthropic, Google Gemini, and DeepSeek are LLM-only (pair them with an OpenAI-compatible embedder). DeepSeek is the most cost-effective option (~$0.001-0.005/page).</details>@@ -243,6 +247,7 @@ Hyper-Extract uses LangChain structured output with **function calling**. The mo|----------|-----------------|| **OpenAI** | gpt-4o, gpt-4o-mini, gpt-5 || **Anthropic** | claude-opus-4-8, claude-sonnet-4-6, claude-haiku-4-5 |+| **Google Gemini** | gemini-3.8-flash || **DeepSeek** | deepseek-v4-flash, deepseek-v4-pro || **阿里云百炼** | qwen-plus, qwen-turbo, deepseek-r1 || **Local vLLM** | Qwen3.5-9B (GPTQ-Marlin) |@@ -250,7 +255,7 @@ Hyper-Extract uses LangChain structured output with **function calling**. The mo**Embedding models** (semantic search) work with any OpenAI-compatible endpoint: `text-embedding-3-small`, `text-embedding-v4` (Bailian), `bge-m3` (local vLLM).<details>-<summary><b>Provider notes</b> — DeepSeek & Anthropic pairing</summary>+<summary><b>Provider notes</b> — DeepSeek, Anthropic & Gemini pairing</summary><br>> **DeepSeek:** V4 models default to "thinking" mode, which Hyper-Extract auto-disables so structured extraction works. Set `DEEPSEEK_API_KEY`. DeepSeek has no embeddings API:@@ -267,6 +272,13 @@ Hyper-Extract uses LangChain structured output with **function calling**. The mo> llm, emb = create_client(llm="anthropic", embedder="openai:text-embedding-3-small")> ```+> **Google Gemini:** Gemini is used for the **LLM** (set `GOOGLE_API_KEY` or `GEMINI_API_KEY`, extra: `pip install 'hyperextract[google]'`). Default model is `gemini-3.8-flash`. No mature embedder path in this repo:+>+> ```python+> from hyperextract import create_client+> llm, emb = create_client(llm="google", embedder="openai:text-embedding-3-small")+> ```+</details>> 📖 Full guide: [Provider System & Local Model Support](https://yifanfeng97.github.io/Hyper-Extract/latest/concepts/provider-system/)diff --git a/README_ZH.md b/README_ZH.mdindex 2eb25de..8ff6e50 100644--- a/README_ZH.md+++ b/README_ZH.md@@ -83,7 +83,7 @@ he config embedder -p vllm -u http://localhost:8001/v1 -k dummy -m BAAI/bge-m3```<details>-<summary><b>更多 provider</b> — Anthropic (Claude)、阿里云百炼、OrcaRouter…</summary>+<summary><b>更多 provider</b> — Anthropic (Claude)、Google Gemini、阿里云百炼、OrcaRouter…</summary><br>```bash@@ -91,11 +91,15 @@ he config embedder -p vllm -u http://localhost:8001/v1 -k dummy -m BAAI/bge-m3he config llm -p anthropic -k YOUR_ANTHROPIC_API_KEYhe config embedder -p openai -k YOUR_OPENAI_API_KEY+# Google Gemini —— 仅 LLM(默认:gemini-3.8-flash)+he config llm -p google -k YOUR_GOOGLE_API_KEY+he config embedder -p openai -k YOUR_OPENAI_API_KEY+# 百炼(阿里云,Qwen,一个 key 同时提供 LLM 和 embedding)he config init -p bailian -k YOUR_BAILIAN_API_KEY```-> OpenAI 和百炼同时提供 LLM 和 embedding 模型;Anthropic 和 DeepSeek 仅提供 LLM(需搭配 OpenAI 兼容 embedder)。DeepSeek 最经济(约 $0.001-0.005/页)。+> OpenAI 和百炼同时提供 LLM 和 embedding 模型;Anthropic、Google Gemini 和 DeepSeek 仅提供 LLM(需搭配 OpenAI 兼容 embedder)。DeepSeek 最经济(约 $0.001-0.005/页)。</details>@@ -243,6 +247,7 @@ Hyper-Extract 通过 LangChain 结构化输出的 **function calling** 方法工|----------|-----------------|| **OpenAI** | gpt-4o, gpt-4o-mini, gpt-5 || **Anthropic** | claude-opus-4-8, claude-sonnet-4-6, claude-haiku-4-5 |+| **Google Gemini** | gemini-3.8-flash || **DeepSeek** | deepseek-v4-flash, deepseek-v4-pro || **阿里云百炼** | qwen-plus, qwen-turbo, deepseek-r1 || **本地 vLLM** | Qwen3.5-9B (GPTQ-Marlin) |@@ -250,7 +255,7 @@ Hyper-Extract 通过 LangChain 结构化输出的 **function calling** 方法工**嵌入模型**(语义搜索)支持任意 OpenAI 兼容端点:`text-embedding-3-small`、`text-embedding-v4`(百炼)、`bge-m3`(本地 vLLM)。<details>-<summary><b>Provider 说明</b> — DeepSeek 与 Anthropic 的搭配方式</summary>+<summary><b>Provider 说明</b> — DeepSeek、Anthropic 与 Gemini 的搭配方式</summary><br>> **DeepSeek:** V4 模型默认开启 "thinking" 模式,Hyper-Extract 会自动关闭以保证结构化抽取可用。设置 `DEEPSEEK_API_KEY`。DeepSeek 没有嵌入接口:@@ -267,6 +272,13 @@ Hyper-Extract 通过 LangChain 结构化输出的 **function calling** 方法工> llm, emb = create_client(llm="anthropic", embedder="openai:text-embedding-3-small")> ```+> **Google Gemini:** Gemini 仅用于 **LLM**(设置 `GOOGLE_API_KEY` 或 `GEMINI_API_KEY`,额外依赖:`pip install 'hyperextract[google]'`)。默认模型为 `gemini-3.8-flash`。本仓库没有成熟的 embedder 路径:+>+> ```python+> from hyperextract import create_client+> llm, emb = create_client(llm="google", embedder="openai:text-embedding-3-small")+> ```+</details>> 📖 完整指南:[Provider 系统与本地模型支持](https://yifanfeng97.github.io/Hyper-Extract/latest/zh/concepts/provider-system/)diff --git a/examples/README.md b/examples/README.mdindex 1b7aa4f..539c3b5 100644--- a/examples/README.md+++ b/examples/README.md@@ -21,6 +21,9 @@ python providers/deepseek_demo.py# Anthropicpython providers/anthropic_demo.py+# Google Gemini+python providers/google_demo.py+# OrcaRouterpython providers/orcarouter_demo.py@@ -50,6 +53,7 @@ examples/│ ├── bailian_demo.py # Bailian (Alibaba Cloud) setup│ ├── deepseek_demo.py # DeepSeek setup│ ├── anthropic_demo.py # Anthropic setup+│ ├── google_demo.py # Google Gemini setup│ ├── orcarouter_demo.py # OrcaRouter setup│ └── vllm_demo.py # Local vLLM setup├── en/ # English demosdiff --git a/examples/README_ZH.md b/examples/README_ZH.mdindex cca4946..14ab7ca 100644--- a/examples/README_ZH.md+++ b/examples/README_ZH.md@@ -21,6 +21,9 @@ python providers/deepseek_demo.py# Anthropicpython providers/anthropic_demo.py+# Google Gemini+python providers/google_demo.py+# OrcaRouterpython providers/orcarouter_demo.py@@ -51,6 +54,7 @@ examples/│ ├── bailian_demo.py # 百炼(阿里云)配置│ ├── deepseek_demo.py # DeepSeek 配置│ ├── anthropic_demo.py # Anthropic 配置+│ ├── google_demo.py # Google Gemini 配置│ ├── orcarouter_demo.py # OrcaRouter 配置│ └── vllm_demo.py # 本地 vLLM 配置├── en/ # 英文演示diff --git a/examples/providers/google_demo.py b/examples/providers/google_demo.pynew file mode 100644index 0000000..35cb288--- /dev/null+++ b/examples/providers/google_demo.py@@ -0,0 +1,36 @@+"""+Google Gemini Provider Demo++Extract entities and relationships using the Gemini Developer API.+Gemini has no mature embedder path in this repo, so pair it with an+OpenAI-compatible embedder. Requires `pip install 'hyperextract[google]'`.+Set GOOGLE_API_KEY or GEMINI_API_KEY. No real key is required to read+this file.++Usage:+ GOOGLE_API_KEY=xxx OPENAI_API_KEY=sk-xxx python examples/providers/google_demo.py+"""++from hyperextract import create_client, AutoGraph++# Gemini for LLM, OpenAI for embeddings (no Gemini embedder path here)+llm, emb = create_client(+ llm="google", # default: gemini-3.8-flash+ embedder="openai:text-embedding-3-small",+)++graph = AutoGraph(+ instruction="Extract people and their relationships",+ llm_client=llm,+ embedder=emb,+ node_key_extractor=lambda n: n.name,+ edge_key_extractor=lambda e: (e.source, e.target, e.type),+ nodes_in_edge_extractor=lambda e: (e.source, e.target),+)++text = "Zhang San founded ByteDance. Li Si serves as CEO."+graph.parse(text)++print(f"Nodes: {len(graph.nodes)}, Edges: {len(graph.edges)}")+for n in graph.nodes:+ print(f" - {n.name} ({n.type})")