Official Implementation for "Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flow"
Open accessAbstract
This artifact is the official open-source implementation for ASPLOS 2025 paper "Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flow". It contains the simulator and the prototype system used in this paper. A detailed step-by-step guide to environment setup and example usage is in the readme file.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Helix is a high-throughput serving system for distributed clusters.
“Helix [14]. It is a high-throughput serving system for distributed clusters.”
From Distributed LLM Serving on Consumer-Grade GPUs by Reconciling Computation and Communication (2025) · cites this paper as [14]
The Zenodo Description (no page given) identifies this artifact as the official implementation of Helix, a large-language-model serving system, and says it contains the prototype system. The file listing includes distributed-cluster configurations and throughput plots, but the available passages do not establish that the system achieves high throughput. (LLM, conf=0.78)
Experimental — accuracy not yet measuredExisting distributed serving systems suffer from transmission competition between prefill and decode requests.
“However, existing systems [28, 14, 4, 12] suffer from transmission competition between requests at different phases.”
From Distributed LLM Serving on Consumer-Grade GPUs by Reconciling Computation and Communication (2025) · cites this paper as [14]
The passages from our copy are repository file listings, with no section or page labels. Searches for transmission competition between phases and for prefill–decode network contention found no passage addressing whether existing systems suffer from it. (LLM, conf=0.72)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cited by
Sorted by citations
Distributed LLM Serving on Consumer-Grade GPUs by Reconciling Computation and Communication
Jin, L., Zhang, K., Chen, Y., Zhuoyifan, Li, R., Gao, Y., Yang, B., Cai, Z., & Dong, W. (2025) · Findings of the Association for Computational Linguistics: EMNLP 2025 · doi:10.18653/v1/2025.findings-emnlp.957
- Authors
- Mei, Y., Zhuang, Y., Miao, X., Yang, J., Jia, Z., & Vinayak, R.
- Affiliations
- Carnegie Mellon University (US)
- Venue
- Zenodo
- Publisher
- Zenodo
- Published
- 2024-11-08
- OpenAlex
- W7223744381
- Cited by
- 0
- References
- 0
- Type
- software
- Language
- en
- Open access
- green
- doi.org
- Locations
- Zenodo (CERN European Organization for Nuclear Research)
- Topics
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Computer Science › Information Systems › Cloud Computing and Resource Management
- Physical Sciences › Computer Science › Computer Networks and Communications › Distributed and Parallel Computing Systems
- Keywords
- large language model serving, heterogeneous networks, maximum flow, load balancing, GPU resource allocation, simulators, prototype system
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
