CoolFace
Modelpublic

hbmartin/creg-sql-xgenerationlab-xiyansql-qwencoder-3b-2502-mlx-4bit

sourceHugging Faceotherupdated 2mo agoView on Hugging Face
0likes171downloads
Model Card

hbmartin/creg-sql-xgenerationlab-xiyansql-qwencoder-3b-2502-mlx-4bit

An MLX 4-bit text-to-SQL derivative for the frozen synthetic CREG commercial real-estate portfolio. This artifact is a research prototype, not a general SQL model.

Reproducibility

  • —Base: XGenerationLab/XiYanSQL-QwenCoder-3B-2502@0d35ce19d4e4d78d82d19b18b3e0f1c7a86eb535
  • —Verified base artifact tree SHA-256: a527bbfcb01aeb07017483f3204f6c32df1654dfd70f019b1f624ede002afe23
  • —Base artifact lock SHA-256: 555cf1b23296de5321f068e1e02c4509cef9b6e691adf15a36401f831a7dba26
  • —Code revision: a35f6c06ad031e91512ba3977c3430d241b59493 (dirty state: true)
  • —Training run: qlora-xiyansql-qwencoder-3b-seed-424242
  • —Training runner SHA-256: b2765239ade40bd3231f057de1aec431ac227e7a8ce9769fc4651df29303943c
  • —Corpus generator SHA-256: 416f1370045de5fffe69a557fa1486fd80f4de645a7f4311b9ae878e3ccd79c9
  • —Model manifest input SHA-256: 3c5c293da1760ea054862d35ddd2e231f458a5bfec8d1386e5fd4bb72044fb9c
  • —Pinned Python lock SHA-256: 034b816d176c52f6a0ba5eb5f657f1dc1a7fbd222edd564e32e75361adb91cd3
  • —Training configuration SHA-256: 63a139bb79e5ad77d0241b9b76d0c2892a3459019bcfad04367d8d932800b545
  • —Corpus manifest SHA-256: 4e3d25923a8426ba3c09349d240a1c77aca79daf5c69631694d28dfe4374cabe
  • —Gold set remained held out: 2bde4dedc23bc7938d0250f2e2d5e22502903c8caad6c0951e1bbca9d0d77036
  • —Adapter tree SHA-256: 8726d3bd0594661dd077907b6d463da028f9856e591e717eeca1564bbe8e2efb
  • —Training log SHA-256: ac63718348727ecd2ad62fdba27b5c9475f17e823f1e4c4991873511074f8fe5
  • —Fused output tree SHA-256 before publication documentation: af4ec6faa63ef2dbae8731004f61e9d621c256ca826a7aa84451eabe239a0840
  • —Model payload SHA-256 excluding documentation, license, and notice files: 1c948fb0ead2a55a76258398ce0dd99fc6e0dd341b160e46c31500153d263025
  • —Quantization: fused 4-bit affine, group size 64
  • —Commercial use allowed by the declared inherited license: false

Training corpus inputs:

  • —fine-tuning/synth/out/train.jsonl: 3a9ad4806692cdc89e8e68c77e29c5e1eedaefac5745c3a87bd4e4fb1758021e (byte-for-byte regeneration: true)
  • —fine-tuning/synth/out/valid.jsonl: b0e72fde78f50e80bc5bdd4664eb7e87695db8a52e042067cfb32ddfc0fcec33 (byte-for-byte regeneration: true)
  • —fine-tuning/synth/out/gate_stats.json: d63b2ae38ec22276dc7706c34d9f12b5f301909e50f01c3c7412043abd22f442 (byte-for-byte regeneration: true)

The complete YAML configuration uses seed 424242, 600 iterations, batch size 4, 16 adapted layers, learning rate 1e-4, prompt masking, and explicit mlx-lm defaults. The immutable training run retains the complete commands, per-file adapter inventory, training log, and fused output inventory.

yaml
# Complete replacement for the incompletely recorded PR #1 training command.
# The finalist runner must override `model` and `adapter_path`; sentinel values
# make a missing override fail rather than selecting an implicit upstream.
model: REQUIRED_FINALIST_MODEL_OVERRIDE
train: true
fine_tune_type: lora
optimizer: adam
optimizer_config:
  adam: {}
  adamw: {}
  muon: {}
  sgd: {}
  adafactor: {}
data: synth/out
seed: 424242
num_layers: 16
batch_size: 4
iters: 600
val_batches: 25
learning_rate: 0.0001
steps_per_report: 10
steps_per_eval: 200
grad_accumulation_steps: 1
resume_adapter_file: null
adapter_path: REQUIRED_IMMUTABLE_ADAPTER_PATH_OVERRIDE
save_every: 100
test: false
test_batches: 500
max_seq_length: 2048
config: null
grad_checkpoint: false
clear_cache_threshold: 0
lr_schedule: null
lora_parameters:
  rank: 8
  dropout: 0.0
  scale: 20.0
mask_prompt: true
report_to: null
project_name: creg-sql

Evaluation

  • —gold_v2.jsonl; GCD on; temperature 0.0; seed 0: EX 0.655, valid SQL 0.930, p95 3008954 μs; immutable run matrix-fine-tune-gold-v2-ft-xiyansql-qwencoder-3b-gcd-on-t-0_0-s-0 (manifest SHA-256 becae97c3393e1f7c8ebbc7a1414afbf95e467a9d0ed50498c03baa802e38954, summary SHA-256 82dfe226d8376069cc2a4ccc6f48551c41f02c74efbea31205a9226e3a0ff3a6)
  • —gold_v2.jsonl; GCD off; temperature 0.0; seed 0: EX 0.650, valid SQL 0.925, p95 1559472 μs; immutable run matrix-fine-tune-gold-v2-ft-xiyansql-qwencoder-3b-gcd-off-t-0_0-s-0 (manifest SHA-256 83850b41bb1ec00343e485f5a7e3fbd092785b60edcd78f68cb7ed98354b9b77, summary SHA-256 c7a73729ea1855c26b89ae7fefb42f01372436efc3ddce10fa43e1dc26c6f348)

Execution accuracy is order-insensitive typed row-multiset equality with four-decimal half-even numeric normalization. These scores apply only to the frozen CREG schema/database/gold set and their immutable run manifests.

Limitations

  • —Narrow synthetic domain and fixed SQLite schema.
  • —May generate semantically incorrect, incomplete, or non-executable SQL.
  • —Not evaluated for arbitrary databases, adversarial prompts, or production financial decision-making.
  • —Generated SQL must execute under a read-only connection and should be independently reviewed.

License and required notice

This is a modified derivative of Qwen2.5-Coder-3B-Instruct. It is provided under the Qwen Research License included in this repository, together with any additional upstream license file identified by the base artifact. Non-commercial use only. Built/Improved using Qwen. Qwen, the base-model authors, and any intermediate model authors are attributed through the base-model link, NOTICE, modification notice, and included license files.