Qwen3-4B File-Ops Distilled
Summary
This is the first TinyGPT specialist package for a model we actually built:
a fused Qwen3-4B-Instruct-2507 bf16 HF/MLX safetensors directory distilled for
GorillaFileSystem multi-turn file-operation tasks.
It is a routed specialist, not the general Pace planner.
Artifact
- Package id:
qwen3-4b-file-ops-distilled
- Public artifact:
sarthakagrawal927/qwen3-4b-file-ops-distilled
- Public storage target: Hugging Face Hub model repo
- Format: HF/MLX safetensors directory
- Base:
Qwen/Qwen3-4B-Instruct-2507
- Precision: bf16
- Training method: frontier/gold trajectory distillation rendered in the
student's native tool-calling chat template
Measured Result
| Suite | Stock 4B | Distilled 4B |
|---|
| File-ops hard gate | 58% | 100% |
| File-ops hardgen held-out | - | 95% |
| Out-of-domain breadth | 59.6% | 42.3% |
The file-ops domain saturated at 4B: the distilled model matched frontier on
the hard and veryhard file-ops gates. The same training caused negative
transfer outside that domain, so this package is only correct behind a router.
Recommended Use
Use this model when the router has already identified a file-operation task
with derivable arguments: paths, file names, directories, moves, creates,
deletes, and navigation through a file-system backend.
Do not use it as a general planner. For general multi-domain planning, use the
planner lock in docs/planner-lock-2026-06-19.md: stock
Qwen3-4B-Instruct-2507 bf16 with the plan-then-execute prompt.
Known Limits
- The specialist regresses on non-filesystem BFCL multi-turn backends.
- It was not validated as a broad Pace planner.
- It depends on correct routing. Bad routing turns a narrow win into a broad
regression.
- The artifact is multi-GB and is published on Hugging Face Hub, not committed
to this repository or required to remain in local cache.
References
docs/learn/tool-calling-frontier-parity.md sections 8.1-8.5docs/planner-lock-2026-06-19.md
