gfm-rag
[NeurIPS'25, ICLR'26] Graph Foundation Model for Retrieval Augmented Generation
File Explorer
Download Latest Version (.zip)- code-quality.yml
- gh-pages.yml
- python-publish.yml
- el_model.md
- graph_constructor.md
- ner_model.md
- openie_model.md
- sft_constructor.md
- datasets.md
- doc_ranker.md
- evaluator.md
- gfmrag_retriever.md
- graph_indexer.md
- llms.md
- losses.md
- models.md
- prompt_builder.md
- text_emb_models.md
- doc_ranker_config.md
- el_model_config.md
- g_reasoner_config.md
- g_reasoner_graph_index_config.md
- g_reasoner_retrieval_config.md
- g_reasoner_sft_config.md
- gfm_rag_config.md
- gfm_rag_graph_index_config.md
- gfmrag_finetune_config.md
- gfmrag_pretrain_config.md
- gfmrag_retriever_config.md
- graph_constructor_config.md
- index.md
- kg_index_config.md
- ner_model_config.md
- openie_model_config.md
- sft_constructor_config.md
- text_embedding_config.md
- wandb_config.md
- g_reasoner.md
- gfm_rag.md
- overview.md
- quick_start.md
- g-reasoner.png
- intro.png
- data_format.md
- index.md
- path_interpretation.md
- retrieval_and_qa.md
- training.md
- CHANGELOG.md
- DEVELOPING.md
- index.md
- install.md
- __init__.py
- base_evaluator.py
- hotpot_qa_evaluator.py
- musique_evaluator.py
- retrieval_evaluator.py
- two_wiki_qa_evaluator.py
- __init__.py
- base_model.py
- colbert_el_model.py
- dpr_el_model.py
- __init__.py
- base_graph_constructor.py
- kg_constructor.py
- __init__.py
- base_model.py
- llm_ner_model.py
- __init__.py
- base_model.py
- llm_openie_model.py
- __init__.py
- dspy_prompts.json
- rerank.py
- __init__.py
- base_sft_constructor.py
- gfm_rag_constructor.py
- gfm_reasoner_constructor.py
- hipporag2_constructor.py
- __init__.py
- langchain_util.py
- openie_extraction_instructions.py
- utils.py
- __init__.py
- graph_dataset_loader.py
- graph_index_dataset.py
- graph_index_dataset_v1.py
- __init__.py
- base_hf_causal_model.py
- base_language_model.py
- chatgpt.py
- __init__.py
- model.py
- rankers.py
- __init__.py
- model.py
- operator.cuh
- rspmm.cpp
- rspmm.cu
- rspmm.h
- util.cuh
- __init__.py
- rspmm.py
- __init__.py
- base_nbfnet.py
- layers.py
- models.py
- query_utils.py
- tasks.py
- util.py
- variadic.py
- __init__.py
- base_model.py
- __init__.py
- base_model.py
- configuration_nvembed.py
- modeling_nvembed.py
- nv_embed.py
- qwen3_model.py
- __init__.py
- base_trainer.py
- kgc_trainer.py
- sft_trainer.py
- training_args.py
- __init__.py
- dist_graph_utils.py
- qa_utils.py
- setup_training.py
- util.py
- wandb_utils.py
- 2wikimultihopqa_ircot.yaml
- hotpotqa_ircot.yaml
- musique_ircot.yaml
- doc_ent_ranker.yaml
- idf_ranker.yaml
- idf_topk_ranker.yaml
- simple_ranker.yaml
- topk_ranker.yaml
- colbert_el_model.yaml
- dpr_el_model.yaml
- gte_large_en_1.5.yaml
- gte_qwen2_1.5b.yaml
- gte_qwen2_7b.yaml
- nv_embed_v2.yaml
- exp_visualize_path.yaml
- index_dataset.yaml
- kgc_training.yaml
- qa_inference.yaml
- qa_ircot_inference.yaml
- sft_training.yaml
- visualize_path.yaml
- index_dataset.yaml
- kgc_trianing.yaml
- qa_inference.yaml
- sft_training.yaml
- sft_training_w_answer.yaml
- stage3_qa_ircot_inference.yaml
- visualize_path.yaml
- kg_constructor.yaml
- llm_ner_model.yaml
- llm_openie_model.yaml
- 2wikimultihopqa.yaml
- hotpotqa.yaml
- musique.yaml
- 2wikimultihopqa.yaml
- hotpotqa.yaml
- musique.yaml
- zero_shot.yaml
- gfm_rag_sft_constructor.yaml
- gfm_reasoner_sft_constructor.yaml
- hipporag2_sft_constructor.yaml
- bge_large_en.yaml
- gte_large_en_1.5.yaml
- gte_qwen2_1.5b.yaml
- gte_qwen2_7b.yaml
- mpnet.yaml
- nv_embed_v2.yaml
- qwen3.yaml
- qwen3_8b.yaml
- default.yaml
- __init__.py
- visualize_path.py
- __init__.py
- index_dataset.py
- kgc_training.py
- qa.py
- qa_ircot_inference.py
- README.md
- sft_training.py
- __init__.py
- gfmrag_retriever.py
- graph_indexer.py
- losses.py
- prompt_builder.py
- stage1_data_index.sh
- stage2_evaluate.sh
- stage2_finetune.sh
- stage3_qa_inference.sh
- stage3_qa_ircot_inference.sh
- stage1_data_index.sh
- stage2_finetune.sh
- stage2_pretrain.sh
- stage3_qa_inference.sh
- stage3_qa_ircot_inference.sh
- fb.yaml
- mc.yaml
- ms.yaml
- oe.yaml
- tf.yaml
- complex_reasoning.yaml
- contextual_summary.yaml
- creative_generation.yaml
- fact_retrieval.yaml
- complex_reasoning.yaml
- contextual_summary.yaml
- creative_generation.yaml
- fact_retrieval.yaml
- graphrag_bench_cs_qa_inference.yaml
- graphrag_benchmark_qa_inference.yaml
- __init__.py
- graphrag_bench_qa.py
- graphrag_benchmark_qa.py
- prompt_utils.py
- stage1_data_index.sh
- stage2_retrieval.sh
- stage3_qa_inference_graphrag_bench.sh
- stage3_qa_inference_graphrag_benchmark.sh
- README.md
- __init__.py
- profile_split_graph.py
- test_el_model.py
- test_gfm_rag_scripts.py
- test_gfmrag_retriever.py
- test_ner_model.py
- test_ner_model_parallel.py
- test_openie_model.py
- test_openie_model_parallel.py
- test_pretrained_dataset_config.py
- test_stage2_dataset.py
- test_text_emb_model.py
- .env.example
- .gitignore
- .pre-commit-config.yaml
- LICENSE
- mkdocs.yml
- mypy.ini
- poetry.lock
- pyproject.toml
- pytest.ini
- README.md
# Installation Guide
1. Get the code
git clone https://github.com/RManLuo/gfm-rag
Downloads the entire project code from GitHub to your computer.
cd gfm-rag
Moves into the project folder you just downloaded.
2. Official Install Script
Easy RecommendedPrerequisites
- Python 3 Python is required to use pip.
pip install gfmrag
Installs the package published on PyPI directly β no need to clone the source.
After installing, open a new terminal and run the program's version command (e.g. --version) to confirm it worked.
Pulled directly from this repo's README.
3. Python
EasyPrerequisites
pip install gfmrag
Installs the package published on PyPI directly β no need to clone the source.
python -m gfmrag.workflow.sft_training --config-path config/gfm_reasoner
Runs the Python script (or module).
If it runs without errors and prints output in the terminal, it worked.
Pulled directly from this repo's README.
// repository documentation
Was this content helpful?
(0 ratings)