ALPHABELL RESEARCH
Post-training, model behavior and systems that improve AI.
Change one word, guess the precision, scale a feed-forward branch, follow corrected facts, or hear real routing as music. Explore 60 study generations from Flint and alphaBell Qwen Open, with downloadable evidence and exact CPU reproduction.
Real captures, blind comparisons, activation surgery and model music.
Draw text, object and character boxes for MOSS-Image-Base. Keep exact copy in an editable draft, then export the centered model-coordinate JSON.
Runs locally in your browser.
Inspect eight saved prompt-format probes for DeepSeek-V4.1-Flash. Compare raw and normalized tool history, numeric reasoning settings, and exact Python/Jinja outputs.
Saved cases with exact inputs and pinned source references.
Three real Flint coding failures. Edit the source, inspect the browser error, reveal the recorded repair and play the result.
Real before/after source and hashes. Runs locally in your browser, with no inference call.
See which settings Rec's Flint MLX server accepts, rejects or silently ignores. Review JSON locally and trace every diagnostic to pinned source.
Describe a strange game, play it, then give it one more twist. Stock Flint Q8_0 adapts a tested game foundation, with isolated browser checks and bounded repair.
Each of the three curated games has a recorded winning playthrough using real controls. Explore the code and the generation attempts, including failures.
Application and alphaBell research data under Apache 2.0. Stock model with browser feedback and repair.
Explore our Qwen3.8-27B research checkpoint, historical coding comparisons and evaluation records.
Noncommercial research. See the model card for lineage, results and release terms.