REVIEW 12 cited by
Inference economics of language models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Inference economics of language models
read the original abstract
We develop a theoretical model that addresses the economic trade-off between cost per token versus serial token generation speed when deploying LLMs for inference at scale. Our model takes into account arithmetic, memory bandwidth, network bandwidth and latency constraints; and optimizes over different parallelism setups and batch sizes to find the ones that optimize serial inference speed at a given cost per token. We use the model to compute Pareto frontiers of serial speed versus cost per token for popular language models.
Forward citations
Cited by 12 Pith papers
-
Not All Errors Are Equal: A Systematic Study of Error Propagation in Large Language Model Inference
A new fault-injection framework enables a systematic empirical study that produces 17 takeaways on error propagation in LLM inference and four software-only mitigation directions.
-
Analyzing Reverse Address Translation Overheads in Multi-GPU Scale-Up Pods
Simulation study shows cold TLB misses in reverse address translation dominate latency for small collectives in multi-GPU pods, causing up to 1.4x degradation, while larger ones see diminishing returns.
-
Think Before You Grid-Search: Floor-First Triage for LLM Serving
LLM serving should triage by five-resource analytical floors and wall ordering, not grid search; on 16×H20, TP16 is capacity-capped at ~70 while EP+DP attention reaches ~644 concurrent 8K requests.
-
Think Before You Grid-Search: Floor-First Triage for LLM Serving
A five-dimensional resource-vector floor model computes latency bounds and capacity walls for LLM serving, predicting when TP16 or EP16+DP attention layouts dominate based on operating point.
-
Beyond Per-Token Pricing: A Concurrency-Aware Methodology for LLM Infrastructure Cost Estimation
Effective LLM inference cost per million output tokens varies 2.5-36x with offered request rate due to utilization, addressed by a concurrency-aware measurement methodology and open-source vLLM tool validated across m...
-
Scale Weight Decay and Train Better
Muon with weight decay scaled by η/η_max reaches the same MoE validation loss ~30% faster than constant-decay Muon while preserving asymptotic stationarity of the unregularized objective.
-
Efficient Clustering with Provable Guardrails for LLM Inference at Scale
Mini-Batch K-Means followed by greedy set-cover within each bucket guarantees every sample lands with a representative that is at least α-similar and attribute-identical, reducing LLM inference cost ~50× at 38M-custom...
-
The Productivity-Reliability Paradox: Specification-Driven Governance for AI-Augmented Software Development
The Productivity-Reliability Paradox arises because AI code generators produce variable output while developers lack sufficient specification discipline, making governance models focused on specifications the binding ...
-
A Techno-Economic Framework for Cost Modeling and Revenue Opportunities in Open and Programmable AI-RAN
Techno-economic framework shows that GPU AI-RAN deployments can offset extra costs via AI revenue for up to 8x ROI across scenarios with varying token depreciation, demand, and GPU densities.
-
A Techno-Economic Framework for Cost Modeling and Revenue Opportunities in Open and Programmable AI-RAN
GPU-accelerated RAN can achieve up to 8x ROI by leasing surplus compute capacity to AI tenants, offsetting added capital and operational costs.
-
Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches
Post-training N:M activation pruning preserves generative performance in LLMs better than equivalent weight pruning, with the 8:16 pattern emerging as a practical hardware-friendly choice.
-
Tokalator: A Context Engineering Toolkit for Artificial Intelligence Coding Assistants
Tokalator is a toolkit with VS Code extension, calculators, and community resources to monitor and optimize token usage in AI coding environments.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.