Open Model Lab

Open Model Lab Timeline

Completed dates stay fixed; future stages follow actual model/data prerequisites. Development and local compute proceed independently, without a replacement finish-date promise.

Completed evidence and next artifact gates

Artifact-gated
Artifact gate

Eval Suite v2

Repeated runs reproduce their scores, and per-category 95% confidence intervals are ±5 points or narrower.

planned
After v2 base runs
Artifact gate

SFT Pipeline + Data Quality

The work does not stop at fine-tuning; it shows measurable behavior changes, including regressions, beyond eval suite v2's confidence intervals.

planned
After SFT checkpoint
Artifact gate

Preference Optimization / DPO

The report clearly shows where post-training helps and where it creates risk.

planned
After usable tool interfaces
Artifact gate

Agent Harness / Tool Use

Agent behavior is measured step by step with failure modes, not only success/failure.

planned
After Phase 1 evidence
Artifact gate

Release 1 — Phase 1 Portfolio

An external reader can reproduce one stage's main result from the repository alone.

planned
After reasoning checkpoints
Artifact gate

Reasoning Behavior + Process Evaluation

The system can show flawed processes despite correct answers, or identify breakpoints leading to wrong answers.

planned