> Canonical page: https://posttrainllm.com/artifacts/qwen3-4b-rest-fused

Routed depth and safety win Release-ready weights 2026-09-03

# Qwen3-4B ReST Fused

On the frozen paired run, ReST improved file-ops depth from 9/12 to 12/12, removed eight unexpected side effects, and completed that slice 2.42x faster wall-clock. The same candidate fell from 30/45 to 25/45 on frontier-qualified breadth. The honest win is routed specialization, not a general-planner claim.

- agentic
- ReST
- tool calling
- BFCL

## 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.

File-ops hard gate

 12/12 vs stock 9/12 measured

Qualified breadth

 25/45 vs stock 30/45 measured

Depth wall speedup

 2.42x 360.50s → 148.91s derived

Safety events removed

 10 11 total → 1 across both gates measured

## Competitive Context

**System**

**Metric**

**Score**

**Size / Class**

**Comparable?**

**Readout**

 posttrainllm Qwen3-4B ReST file-ops depth 100% 4B, 8.06GB stored Direct Fresh run: 12/12, zero unexpected side effects, and 13.40 tok/s.

 Stock Qwen3-4B same file-ops depth fixture 75% 4B Direct Fresh run: 9/12, eight unexpected side effects, and 12.10 tok/s.

 ReST breadth result frontier-qualified breadth 55.6% 4B Direct Below stock's 66.7%; this is why the frozen general-successor decision is reject.

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

## Fresh requalification

**Gate**

**Stock**

**ReST candidate**

**Readout**

 File-ops depth 9/12 12/12 +25 points

 Qualified breadth 30/45 25/45 -11.1 points

 Depth wall time 360.50s 148.91s 2.42x faster

 Depth peak RSS 6.77GB 8.13GB +1.36GB

## Release Blockers

### General breadth regressed

The fresh candidate lost five of 45 frontier-qualified breadth cases versus stock.

 Unblock: Do not unblock this candidate as a general model; route it only to the file-ops envelope.

### Not a Pace planner

Pace uses a different intent envelope and its own six-dimension ship gate.

 Unblock: Re-distill on Pace's action surface and clear the Pace gate before runtime wiring.

## 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.

 Shipped — routed only Not fully verified — see the report card for why

[Read the report card →](https://posttrainllm.com/report-cards/qwen3-4b-rest-fused)[Machine-readable JSON](https://posttrainllm.com/report-cards/qwen3-4b-rest-fused.json)

## Evidence

- [Model card](https://github.com/PostTrainLLM/posttrainllm/blob/main/specialists/qwen3-4b-rest-fused/model_card.md)
- [Hugging Face model](https://huggingface.co/posttrainllm/qwen3-4b-rest-fused)
- [Eval report](https://github.com/PostTrainLLM/posttrainllm/blob/main/specialists/qwen3-4b-rest-fused/eval_report.json)
- [Fresh machine-readable result](https://github.com/PostTrainLLM/posttrainllm/blob/main/evals/verified-wins/rest-requalification-result-v1.json)
- [Specialist registry](https://github.com/PostTrainLLM/posttrainllm/blob/main/specialists/registry.json)

## Next Release Action

Keep this package behind the file-ops route and reject it as the general Pace successor.