Bittensor inference gateway Engy overtakes Chutes with 29.3B daily tokens
MarsSmuff · x · 2026-10-12
Engy, an LLM inference gateway in the Bittensor ecosystem, has surpassed Chutes' 7-day average in daily token throughput: Engy served 29.3B tokens yesterday vs Chutes' 20.18B — nearly 50% more. The dashboard shows traffic mostly served by open models like deepseek-v4.1-flash, qwen3.8-27b, kimi-k3 and glm-5.3-flash. The Engy team noted it also has 221 GPUs rented at above-cost hourly rates, shifting from token-maxxing toward compute rental, quipping that LLM inference is a race to zero while GPU prices race to infinity.