You Can Train a 1B Model in 3 Weeks: Dev Shares 15.9-Day Training Run Log
arthurcolle · x · 2026-10-26
Developer arthurcolle says anyone can train a 1B model "if you just chill for 3 weeks and stick to the plan," sharing a live run log: project sys1-999m-v100k-run1 is at 15,940/30,000 steps (53.1%), 2.61B tokens processed, 3,773 tokens/sec, with the process up for 7 days 21 hours and roughly 7 days remaining. The takeaway: small-model pretraining is feasible on a personal time and compute budget if you stay consistent.
More from Infra
- RTX 4090 workstation vs AI subscriptions: an editor's break-even math — iamthecarguy1234 · 2026-10-26
- Cerebras CS-4 teardown: 2x bandwidth and power, 50% fewer parts vs CS-3 — jwt0625 · 2026-10-26
- theo doubts Windows ARM: porting 20 years of tech debt vs Intel improving — wateriscoding · 2026-10-26
- TokenSpeed's Three-Language Stack: C++ for Correctness, Rust for Agents, Python for Glue — zhyncs42 · 2026-10-26
- Speculative Decoding Explained: a 30% Wrong Guesser Never Changes the Output — anbarth · 2026-10-26
- TWIG: weights as shared discrete decision paths unify pruning, quantization, and adder-based circuits — ITMO · 2026-10-26