LingBot-VLA 2.0 — LIBERO full fine-tune, step 1200 (OpenRoboto sim/2)
Canonical parent: openroboto-ai/lingbot-vla-v2-6b-libero @ ce6a322157acc7a03d0ca71bb84423c7f2e124d7 (the season base: LingBot-VLA 2.0 already LIBERO-tuned, itself a fine-tune of robbyant/lingbot-vla-v2-6b@11c703bf).
Method: full-parameter fine-tune with the vendor trainer (Robbyant/lingbot-vla-v2 @ 951475ae)
on the official pool openroboto-ai/libero-datapool @ 2c065c6d (byte-identical mirror of
lerobot/libero @ a1aaacb7; all 40 tasks, 1693 episodes), global batch 64, muon lr 5e-5,
cosine, 1200 optimizer steps, 4x A100-80GB. Normalization stats, camera mapping and 7-dim action
semantics follow the public reference fine-tune unchanged (training/norm_stats.json).
No LIBERO-Pro training samples were used.
- Downloads last month
- 142