nanochat-omni

Author	SHA1	Message	Date
Fam Zheng	c1108ae01f	ci: nanochat smoke on ailab gpu runner smoke / nanochat-smoke (push) Failing after 7s Details - scripts/smoke.sh: end-to-end nanochat pipeline (dataset → tokenizer → tiny d=6 base_train, 50 iters); idempotent, caches under /data/nanochat-smoke (venv, uv-cache, shards). - CN mirrors baked in: aliyun PyPI, sjtu pytorch-wheels/cu128, hf-mirror. - workflow targets gpu label (ailab). drop earlier dummy workflow.	2026-05-05 21:57:22 +01:00
Fam Zheng	b33e6af06c	ci: add dummy workflow targeting ailab runner dummy / hello (push) Successful in 1s Details	2026-05-05 21:37:18 +01:00
mochi	7b29fff2c8	docs: add NanoChat-Omni feasibility research (closes #2 ) 调研内容包括: - 总体技术方案 (Encoder + Projection + LoRA, 三阶段训练) - 硬件对比 (3090/4090/5090/GB10/B40) - 数据集规划 (S1 弱对齐 + S2 指令 + S3 偏好) - 8 周执行计划, 4 周 MVP - 结论: 可行, 建议立即启动	2026-05-05 21:13:25 +01:00
mochi	f278524d9d	chore: add upstream nanochat as submodule at upstream/nanochat Pin to karpathy/nanochat@dc54a1a. Closes #1.	2026-05-05 21:02:46 +01:00
fam	147b21c31b	Initial commit	2026-05-05 20:00:43 +00:00