Harvard open sources LLM inference traces: 6.12 billion requests from a year of production traffic
markjeffrey · x · 2026-09-22
Harvard, in collaboration with Chutes (an open inference network on Bittensor averaging 27B tokens/day), released a paper studying a full year of production LLM traffic: 6.12 billion requests of inference metadata traces, fully open sourced. The dataset supports research on real-world LLM serving workload understanding, system design, and infrastructure optimization.
Related event: Harvard Releases Dataset of 6.12 Billion Real-World LLM Inference Requests(3 posts)→
More from Research
- Autonomous agent edits just 1,024 weights to lift Qwen3.5 0.8B from 7.4% to 37.4% on MATH-500 — VictorSanh · 2026-11-12
- Gait dataset labeled by encounter, spanning pre-op to 3 months post-TKA — zawthet · 2026-11-12
- Walking alignment measured 2.5-3.1° more neutral than X-ray, iPad gait study finds — zawthet · 2026-11-12
- MIT's Tegmark coins 'formal-washing': Signal protocol proof is far from complete — tegmark · 2026-11-12
- Signal formal verification hits 80%, Tegmark says Wired overstated completion — tegmark · 2026-11-12
- Category-theoretic framework asks: when does a physical system actually compute? — neurovium · 2026-11-12