Dream-1 Beta runs robot tasks at $1.72 per success, up to 41x cheaper than LLM agents
gharik · x · 2026-10-20
xdofai (retweeted by gharik) published an API cost comparison of its robot model Dream-1 Beta against four LLM agents on the same tasks: 5–41x cheaper per successful run.
- Dream-1 Beta: $0.25 per attempt, $1.72 per successful run.
- Comparison (per attempt / per successful run): GPT-6 Astra $9.32 / $55.11; Fable 5.1 $6.33 / $69.68; GPT-6.1 Sol $1.79 / $9.41; Opus 5.5 $4.38 / $37.85.
- Methodology: LLMs priced at reported tokens × list price per 1M tokens (robot time excluded); Dream-1 Beta at $0.0023 per action × 22.6 actions/min (staging price list).
Takeaway: a specialized small robot model is far cheaper per action than general-purpose LLM agents, and the gap widens once success rates are factored in.
More from Embodied
- Tactile Curiosity Drives Robot Grasping Without Task Rewards, CoRL 2026 Paper — pabbeel · 2026-10-20
- Dream-1 tops RoboDojo-RC Tier 1 with score of 59, pushing back on 'LLMs will solve robotics' — scott_e_reed · 2026-10-20
- Robotics practitioner: LLM-generated 'robotics startups' all chase the same fake problems — kscottz · 2026-10-20
- Realtime Local AI Video Restyling: 720p 60fps on a Single RTX 5070 Ti — Fit_Revenue1928 · 2026-10-20
- GPT-6 Astra in spatial XR: Jarvis-style live Earth demo on a PICO device — Scobleizer · 2026-10-20
- Long queues at RTX Spark SF launch for local Qwen demos built with ODS software — TheZachMueller · 2026-10-20