miniVERL¶

Auditable single-GPU LLM post-training for choosing, running and inspecting SFT, DPO, knowledge distillation and strict OPD—plus a bounded artifact bridge to one pinned verl profile. miniVERL is independent; no upstream endorsement is implied, and distributed execution is not tested.

Install and run locally Read the compatibility boundary

Install and verify in about a minute¶

Install the PyTorch build that matches your CPU or CUDA system first, then the training extra. This CPU example is deterministic and downloads no model:

python -m pip install torch --index-url https://download.pytorch.org/whl/cpu
python -m pip install "miniverl[train]"
miniverl demo --fast --output runs/quickstart
miniverl inspect runs/quickstart/trajectories.jsonl

The result is a typed trajectory log, checksummed teacher cache, manifest and self-contained report. For CUDA wheels and memory-aware recipes, use the single-GPU guide.

Choose a path¶

Align¶

Choose SFT, DPO, offline KD or OPD from explicit pilot evidence.

miniverl pilot recipes/alignment_tool_policy_toy.yaml --json

Artifact: an Alignment Card with starting checkpoint, metrics, cost and limitations.

Next: When should OPD follow SFT?

Distill locally¶

Use strict OPD, shared-backbone role switching and padded trajectory updates on one CUDA GPU.

miniverl train recipes/qwen_consumer_gpu_shared.yaml --dry-run --json

Artifact: a resolved recipe and typed provenance plan before model loading.

Next: Consumer-GPU shared runtime

Scale out¶

Import only the documented profile, convert Parquet, export standard artifacts and inspect the unsupported boundary.

miniverl bridge doctor exports/my-bundle --json

Artifact: provenance/compatibility-report.json with separate readiness flags; current bundles are not launchable.

Next: verl bridge contract

Measured evidence, kept scoped¶

The Alignment Lab case study starts from an SFT checkpoint already at 100% alignment and 100% retained tool utility on its deterministic sandbox suite. No continuation method improves it; completed regressions remain visible. External IFEval, XSTest, HarmBench and RewardBench endpoints were not executed.

Forest chart of continuation-method alignment and retained-tool-utility percentage-point deltas from the saturated SFT checkpoint, with every seed and mean printed as text.

The consumer runtime result is a systems result, not a new quality claim: shared-backbone role switching and padded trajectory updates reduce measured memory/runtime overhead while preserving the tested local objective. See the Consumer Runtime report and RecoveryBench for full evidence and limits.