Home inference rig serves 1B tokens in its first month, mostly Qwen3.8-Flash
natesiggard · x · 2026-11-11
Developer natesiggard shared his at-home inference setup via Actual: over 1 billion tokens and 110K requests served in the first month. The dashboard shows Qwen3.8-Flash-Next-NV accounted for 1.001B tokens (958M read / 42.8M written), with a small share from Qwen3.8-27b; counts are tallied by the engine on ROBIE with a copy kept on a Pi. A notable real-world local/edge deployment datapoint.
Related event: Home AI Rig Serves Over 1 Billion Tokens in Its First Month(2 posts)→
More from Infra
- Proposal: build data centers underground and use waste heat to melt road snow — MatthewChang · 2026-11-11
- SpaceX to build $24B AI data center and natural gas power plant in Starbase, Texas — Polymarket · 2026-11-11
- Microsoft's DLSS rival Auto SR launches on Xbox Ally X with handheld 1080p and 120fps support — tomwarren · 2026-11-11
- Gemini 4 Carbon lands on Databricks day 0, scoring 66.7% on OfficeQA Pro v2 — matei_zaharia · 2026-11-11
- Anthropic says Claude ported kernels to run on AMD GPUs in just 3 days — scottleibrand · 2026-11-11
- Kimi K3.1 Pro Highspeed hits Amazon Bedrock with 2-3x faster output and 1M-token context — AWS ML Blog · 2026-11-11