#

bird-benchmark

Here are 26 public repositories matching this topic...

khang3004 / AgentSQL-Asym

Production-grade Asymmetric Multi-Agent Text-to-SQL on BIRD-SQL. Offline CHESS/FAISS pruning + MCI-SQL enrichment feed ≤3 Groq API calls: gpt-oss-120b generator · llama-4-scout reflector · gpt-oss-20b critic. LangGraph · SQLAlchemy sandbox · LangSmith · Docker · MIT.

Updated May 24, 2026
Python

agenticfabriq / beacon

Grader and results tracker for text-to-SQL systems: exact-match and got-the-facts grading on BIRD, Spider 1.0 and Spider 2.0-lite, with per-question results behind every published mnemiq figure

python benchmark sql spider leaderboard postgresql grading reproducibility ai-agents text-to-sql nl2sql fastapi llm evals llm-evaluation data-agents bird-benchmark

Updated Oct 25, 2026
Python

craterlabs / struct-sql-distillation

Code for Struct-SQL paper

knowledge-distillation text-to-sql llms bird-benchmark

Updated Mar 11, 2026
Python

shivenkk / sqlpup

From-scratch 400M text-to-SQL language model

nlp sql pytorch language-model fromscratch text-to-sql pretraining llm bird-bench bird-benchmark

Updated Oct 9, 2026
Python

Chaoqi31 / text2sql-agent-rl

Agentic Text-to-SQL trained with RL (GRPO): +7pt execution accuracy on full BIRD-dev from a general Qwen3.5-9B base

reinforcement-learning lora text2sql trl llm vllm qwen grpo bird-benchmark

Updated Jun 24, 2026
Python

brownjuly2003-code / NL_SQL

Portfolio NL→SQL assistant: 94.0% execution accuracy on BIRD Mini-Dev (above the 92.96% human-expert baseline), schema-RAG + sqlglot AST guards + deterministic chart picker, $0 free-tier LLM stack. Streamlit chat UI, live demo. · Портфолио NL→SQL: 94% EA на BIRD Mini-Dev, schema-RAG + sqlglot-гарды + chart picker, бесплатный стек, живое демо.

python portfolio mistral text-to-sql rag fastapi streamlit llm sqlglot nl-to-sql bird-benchmark

Updated Jul 30, 2026
HTML

Miraj-world / Tinysql-coder

Learning-first text-to-SQL fine-tuning with Qwen, LoRA, BIRD, execution evaluation, and guarded SQL repair.

machine-learning sqlite transformers pytorch sql-generation lora peft text-to-sql qwen llm-fine-tuning bird-benchmark execution-evaluation

Updated Jul 15, 2026
Python

Ahsan-20 / Inkling

A privacy-preserving text-to-SQL workspace. Open-weight models on your machine, evaluated against BIRD.

react nlp text-to-sql fastapi typescript-react local-llm ollama local-llm-integration bird-benchmark

Updated Jun 12, 2026
Python

CodeAlekh / sql-analyst-agent

An AI agent that answers questions over unfamiliar SQL databases: one-shot vs agent vs experienced agent on BIRD, with a hand-written Claude tool loop and an MCP database server.

python sqlite mcp ai-agents claude text-to-sql anthropic llm-agents model-context-protocol bird-benchmark

Updated Oct 3, 2026
Python

nkmohit / schema-safe-bench

Reproducible evaluation of schema-grounded Text-to-SQL using BIRD Mini-Dev, with schema retrieval, SQL validation, bounded repair, abstention, and read-only SQLite execution.

benchmark information-retrieval sqlite database-systems natural-language-to-sql text-to-sql ai-systems llm-evaluation llm-safety sql-validation schema-retrieval bird-benchmark

Updated Aug 1, 2026
Python

ivermin1123 / attestql

Audit text-to-SQL gold and predictions on PostgreSQL or SQLite with typed replay evidence.

auditing benchmark sql spider sqlite postgresql evaluation reproducibility text-to-sql bird-benchmark

Updated Oct 22, 2026
Python

Tommyweige / AAP-SQL

Adaptive prompting for Text-to-SQL: selective schema and value augmentation, column retrieval, and relation discovery.

python text-to-sql llm retrieval-augmented-generation schema-linking bird-benchmark

Updated Oct 8, 2026
Python

VineetV2 / Nl2SQL_ATandT_method

Automatic metadata extraction for text-to-SQL — schema linking pipeline reproducing the AT&T method, evaluated on BIRD.

text-to-sql llm schema-linking bird-benchmark

Updated Apr 5, 2026
Python

wenyupapa-sys / nl2sql-draft-candidate-pipeline

Draft SQL and multi-candidate Text-to-SQL pipeline with schema compression, repair, and selector

sql candidate-selection text-to-sql nl2sql schema-linking bird-benchmark

Updated Apr 29, 2026
Python

VineetV2 / Chase_sql

Reproduction of CHASE-SQL (ICLR 2025): divide-and-conquer + query-plan candidate generation with a selection agent, evaluated on the BIRD benchmark.

reproduction text-to-sql llm bird-benchmark

Updated May 21, 2026
Python

garyzava / 4d-metadata-ablation

Reproducibility artifact for the paper 'Does Metadata Completeness Predict Text-to-SQL Accuracy? A Six-Model Ablation on BIRD Mini-Dev'.

metadata reproducibility text-to-sql ablation llm bird-benchmark

Updated Aug 12, 2026
Jupyter Notebook

simenzzz / sextant

An analyst agent that answers questions over a real database — retrieval, AST-validated SQL, bounded repair, uncertainty routing, and full cost accounting, measured against BIRD.

react golang sqlite evaluation ai-agents text-to-sql llm bird-benchmark

Updated Oct 24, 2026
Go

VineetV2 / chase-sql

Reproduction of CHASE-SQL (ICLR 2025): divide-and-conquer + query-plan candidate generation with a selection agent, evaluated on the BIRD benchmark.

reproduction text-to-sql llm bird-benchmark

Updated Mar 31, 2026
Python

d-nyamkhuu / bird-gepa

Optimizing a BIRD text-to-SQL solver's source code with DSPy Flex and GEPA

text-to-sql dspy llm prompt-optimization gepa bird-benchmark

Updated Oct 12, 2026
HTML

shuaihuang028 / execution-guided-text2sql

Execution-guided post-training (QLoRA SFT + DPO) of a 1.5B code LLM for Text-to-SQL on a single 16GB GPU. BIRD dev execution accuracy 20.1% -> 34.4%.

sqlite pytorch lora fine-tuning peft text-to-sql dpo llm qlora bird-benchmark

Updated Oct 18, 2026
Python

Add this topic to your repo

To associate your repository with the bird-benchmark topic, visit your repo's landing page and select "manage topics."