> Canonical page: https://posttrainllm.com/artifacts/qwen06-sql-routed-v1

SQL factory POC Report-ready candidate 2026-07-02

# Qwen3-0.6B Routed SQL Specialist

This is the cleanest current proof of the factory thesis: a tiny 0.6B model can beat a small public SQL baseline on a frozen exact-match slice, but only the routed artifact survives both public and execution-style gates.

- SQL
- routing
- LoRA
- evals

## Headline Numbers

Evidence class applies to each value: measured = retained evaluation or runtime result; historical = recorded result not freshly reproduced; derived = calculated or decided from records; observed = current artifact state; not-measured = an explicit evidence gap.

Public exact

 0.531 64-row b-mc2/sql-create-context slice measured

T5-small baseline

 0.484 same 64 public rows measured

Synthetic execution

 0.860 50 heldout SQLite rows measured

Synthetic exact

 0.840 same 50 heldout rows measured

## Competitive Context

**System**

**Metric**

**Score**

**Size / Class**

**Comparable?**

**Readout**

 posttrainllm routed SQL v1 b-mc2 exact / synthetic exec 0.531 / 0.860 0.6B base + 2 routed LoRAs Direct Current local candidate; public exact and synthetic execution gates are both frozen.

 [T5-small local baseline](https://huggingface.co/cssupport/t5-small-awesome-text-to-sql) b-mc2 exact 0.484 ~60M Direct Same 64-row public slice; posttrainllm is +4.7 points exact on this narrow gate.

 [Defog SQLCoder-7B-2](https://huggingface.co/defog/sqlcoder-7b-2) Defog SQL-Eval category scores 77.1-96% 7B Directional Strong public SQL specialist, but reported as category-level Defog SQL-Eval scores rather than this b-mc2 slice.

 [Snowflake Arctic-Text2SQL-R1-7B](https://www.snowflake.com/en/blog/engineering/arctic-text2sql-r1-sql-generation-benchmark/) BIRD execution accuracy 68.47% 7B Not comparable Useful target class for public SQL execution; posttrainllm must add a BIRD/Spider execution gate before claiming this lane.

 [Snowflake Arctic-Text2SQL-R1-14B / 32B](https://www.snowflake.com/en/blog/engineering/arctic-text2sql-r1-sql-generation-benchmark/) BIRD execution accuracy 70.04% / 71.83% 14B / 32B Not comparable Shows the current public high bar: execution accuracy, not exact string match.

Direct rows share this artifact's eval setup. Directional rows are useful market context but should not be read as leaderboard claims.

## Adapter comparison

**Setup**

**Public exact**

**Synthetic exec**

**Decision**

 Public v4 only 0.531 0.240 Route required

 Blend v1 0.297 0.560 Reject

 Best static composition 0.516 0.460 Reject

 BIRD + b-mc2 v5 0.438 0.280 Reject

 Classifier-routed v1 0.531 0.860 Current best

## Router verification

**Check**

**Result**

**Evidence**

 Unlabeled mixed rows 114 64 public / 50 synthetic

 Public route reason 64 known_public_source

 Synthetic route reason 50 sqlite_db_field

 Route confidence >= 0.99 all smoke rows

## Output-hygiene DPO retries (composed, 50 heldout rows)

**Run**

**Execution**

**Clean-SQL**

**Decision**

 Frozen baseline (SFT) 0.860 0.000 reference

 2026-07-04 ref-free SimPO 0.080 0.000 retry-training (collapse)

 2026-07-11 ref-anchored DPO 0.900 0.000 retry-training

 2026-07-11 higher-pressure DPO 0.920 0.000 retry-data

## Release Blockers

### Public execution benchmark missing

b-mc2 exact match is useful, but serious SQL claims need execution accuracy on public DBs.

 Unblock: Add BIRD Mini-Dev SQLite or Spider SQLite execution fixtures once the DB bundle is local.

### Output hygiene (raw completions carry an Answer:/Explanation: wrapper)

Two reference-anchored DPO retries (2026-07-04 SimPO collapse → fixed; 2026-07-11 gentle + higher-pressure) improved execution 0.860 → 0.900 → 0.920 but left the clean-SQL raw rate at 0.000. The wrapper is a base-model prose prior a rank-4 preference adapter can't strip — the 108 SFT targets are already bare SELECT. Composed DPO is ruled out for hygiene.

 Unblock: Generation-strength fix, not more DPO: stronger SFT (higher rank / more examples) or inference-time steering (constrained SELECT-prefix decoding / stop sequence). Execution is not the problem; only the wrapper is.

### Not a specialist package yet

The adapters currently live under gitignored run folders, not package metadata.

 Unblock: Package under specialists/ only after a ship decision on a public execution gate.

## Fine-Tune Report Card

The portable before/after proof for this artifact: baseline, candidate, deltas, regressions, slices, performance, eval validity, and the ship/retry/reject decision — with every value labeled measured, derived, historical, skipped, or not recorded. Compiled offline from the recorded factory evidence; no model was run to produce it.

 Report only — no model to use Not fully verified — see the report card for why

[Read the report card →](https://posttrainllm.com/report-cards/qwen06-sql-routed-v1)[Machine-readable JSON](https://posttrainllm.com/report-cards/qwen06-sql-routed-v1.json)

## Evidence

- [SQL POC report](https://posttrainllm.com/docs/specialists/b1-sql-poc/)
- [Public artifact registry](https://posttrainllm.com/docs/factory/public-artifacts/)
- [Router smoke](https://github.com/PostTrainLLM/posttrainllm/blob/main/evals/sql-routed-router-smoke.sh)
- [Router implementation](https://github.com/PostTrainLLM/posttrainllm/blob/main/scripts/sql/run_sql_routed_generate.py)

## Next Release Action

Keep the published report artifact report-only. Two hygiene DPO retries proved reference anchoring cures the SimPO collapse and even lifts execution to 0.920, but output hygiene needs a generation-strength fix (stronger SFT or constrained SELECT-prefix decoding), not more preference tuning. Do not present it as a shipped SQL model until public execution and clean-output gates pass.