Topic

Inference Economics

The cost, latency, quality, and verification structure of running AI systems after training.

Why It Matters Here

Inference economics determines whether capability can be used repeatedly, reliably, and affordably.

Programs

Linked Artifacts

Paper / May 16, 2026

Disaggregated or Colocated? The Cost-Frontier of LLM Serving Under SLO Contracts.

Research Paper #1 in the AI systems engineering wedge. A closed-form decomposition of cost per SLO-compliant served token into a prefill term, a decode term, and a KV-transfer tax. Re-derives published throughput from five 2023–2025 systems papers (PagedAttention, Sarathi-Serve, DistServe, Splitwise, Mooncake) into a common frame, plots the first cross-system Pareto frontier under explicit p99 TTFT and p99 TPOT contracts, and solves the break-even surface between colocated and disaggregated architectures. The frontier partitions cleanly.

Paper / May 15, 2026

The Inference-Time Compute Frontier. A Cost-Correct Threshold for Training Versus Test-Time Allocation.

Research Paper #2 in the inference-economics wedge. Derives a closed-form threshold under the Cost-correct decomposition for when the marginal compute dollar reduces cost-per-correct-answer faster on the inference channel than on the training channel. Calibrated against rStar-Math, DeepSeek-R1, and test-time-compute curves; matches the observed frontier-vs-commodity market split.

Paper / May 15, 2026

The Routing Premium. An Economic Threshold for Difficulty-Conditional Inference Compute.

Research Paper #3 in the inference-economics wedge. Derives a closed-form threshold under the Cost-correct decomposition for when conditioning inference compute on a noisy difficulty estimate reduces cost-per-correct-answer: routing pays iff κ·Δ > γ, where κ is classifier calibration, Δ is workload heterogeneity, and γ is classifier overhead. Unifies five published patterns (speculative decoding, cascades, adaptive self-consistency, complexity-aware exploration, early exit) as one allocation rule, and calibrates against six deployed systems with every operating point on the positive side of the threshold.

Field Note / May 16, 2026

Harvesting Serving Slack. ROSE and the Collapsed Train-Serve Boundary.

A daily field note on Gao, Zhao, Muhtar et al.'s ROSE. Cooperative elasticity for agentic RL rollouts on idle serving GPUs. Why the rollout-cost term in Cost-correct can be priced at the marginal-of-idle rate, and what that does to the inference-frontier threshold.

Field Note / May 17, 2026

The Power-Cap Illusion. SM Clock Locking and the Real Decode Lever.

A daily field note on Ma, Afzal, Eitzinger, and Wellein. Power capping does not bite in memory-bound LLM decode on NVIDIA H200. SM clock locking recovers up to 32% of decode energy. Why the standard energy lever moves the wrong knob, and what that does to the decode-cost term in Cost-correct.

Field Note / May 10, 2026

The Verifier as Curriculum. VHG and the Third Role.

A daily field note on Lai, Feng, Teh, and Miao's VHG. Three-party setter-solver-verifier self-play. Why the verifier's job in the production lifecycle just expanded from two places to three.

Field Note / May 3, 2026

The Inference Stack in 2026.

A field note on token economics, runtime systems, model architecture, and the stack changes behind public LLM API price compression.

Definition / Undated

AWQ Quantization

A post-training quantization method that protects high-signal weights using activation statistics.

Definition / Undated

Speculative Decoding

A decoding-time acceleration technique that drafts tokens with a smaller model and verifies them with a larger model.

Definition / Undated

Mamba And State-Space Models

A family of sequence models that replace quadratic attention with selective state-space mechanisms.

Definition / Undated

Edge AI Silicon

The constrained compute layer for running inference close to sensors, robots, drones, and devices.

Definition / Undated

Verification Economics

The study of cost-per-correct-answer and the verifier as a cost-and-value lever in reasoning systems.