Skip to main content
The goal-statement components and prompt blocks are written to work on a frontier model. This page is the rewrite discipline for running the same work on a ~30 GB local model. That means Qwen3.x and GLM-4.x models in the 30-billion-parameter class, in the single-node serving tier.

​
The governing result: shorter is better

The measured result is simple. Cutting the prompt by more than half improved success rate, latency, and cost for agents using models with 30 billion parameters. Restating a model’s default competence wastes tokens and dilutes the task-specific signal. Keep a line only if deleting it changes what the model does. A small model does not fail by ignoring your prompt. It fails by weighting all of it equally. A paragraph of context then costs the same attention as the one hard rule buried inside it. The fix is structural, not rhetorical.

​
Rewrite rules

  1. Short imperative sentences. One idea each. “Run the check. Quote the output.”—not “it would be advisable to validate the result before proceeding.”
  2. One action per line. A numbered checklist where each line is a single tool-sized step. A small model executes checklists reliably; it plans multi-step strategies unreliably.
  3. No implicit reasoning leaps. If step 3 depends on a decision, make the decision a step: “2. If the file exists, go to step 4. If not, go to step 3.” Never assume the model infers the branch.
  4. Hard-coded STOP conditions. Numbers, not judgment: “at most 30 turns,” “never retry a failing tool more than twice,” “three probes with no new evidence -> stop.” A small model cannot be trusted to notice it is looping. The prompt must count for it. Use the T1 hard-stops block verbatim.
  5. Concrete examples over abstractions. Instead of “bound your queries”, write the bounded query: “run | tstats count where index=* by index over the last hour”. One worked example outperforms three sentences of principle.
  6. Verification the model cannot skip. Encode the check in the deliverable’s format, not in an instruction. An evidence table with a mandatory evidence column per row forces a tool call per claim. The model cannot produce the artifact without doing the verification. Prompt-level “please verify” gets skipped; format-level contracts do not.
  7. Bounded choices only. Every unfilled <placeholder> and every “use your judgment” is an unbound choice the small model resolves arbitrarily. Fill placeholders with concrete values; if a choice must stay open, enumerate the options.
  8. Declare failure vocabulary. Give the model the exact words for partial results: BLOCKED, NOT-CHECKED (budget), UNAVAILABLE with the exact error. Without named outcomes, a small model papers over gaps instead of declaring them.
  9. Identity and tools in the variant, behavior in the base. Keep the shared base short and behavioral; per-surface identity, tool lists, and environment go in a small appended variant. They change independently.
  10. Repetition is a decoding failure, not a reasoning failure. Tell the model: output that degrades into repetition means stop generating and flag it. “Thinking harder” through a repetition loop burns the context window and traps the run.

​
What to cut and what to keep

When shrinking a frontier-targeted goal statement to small-model size:

​
Worked example

The same requirement, written for each tier. Frontier-targeted (fine for a large model, unreliable on a 30-billion-parameter model):
Rewrite for 30-billion-parameter models (a template: fill the remaining placeholders before dispatch, per rule 7):
Every judgment call in the first version (“recent,” “actionable,” “well-evidenced,” “avoid duplicates”) became something concrete in the second. Each became a number, an exact query, a named classification, or a format contract.

​
Choosing the tier at all

Prompt tuning does not give a model in the 30-billion-parameter class deep reasoning. Route tasks by what they need. High reasoning depth or cross-source synthesis belongs on the deep-reasoning tier. Long-but-shallow tool chains run fine on that model tier with a checklist prompt. The T2 escalation rubric is the routing contract; choosing a local model holds more on verifying a model before you trust it.
Grade the runs, keep the receipts. Small-model reliability is empirical. Grade each run on completeness, correctness (spot-checked claim rate), evidence quality, and actionability. Use the grading contract block. Then tune the prompt against the grades, not against intuition.