# LLM Serving on Merge dataset Source: [LLM Serving on Merge dataset]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:41:55.162Z Evaluation of LLM Serving performance on the Merge dataset, measured by effective throughput. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | LLM Serving | | Dataset (including any stated split) | Merge dataset | | Metric | Effective Throughput (req/s) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 9 of 9 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Effective Throughput (req/s)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Effective Throughput (req/s) | Result date | Report paper | | --- | --- | --- | --- | | AugServe | 0.6 | Not recorded | 693115d6d6b0dd3342933d87 | | AugServe | 0.57 | Not recorded | 693115d6d6b0dd3342933d87 | | AugServe | 0.51 | Not recorded | 693115d6d6b0dd3342933d87 | | InferCept | 0.27 | Not recorded | 693115d6d6b0dd3342933d87 | | vLLM | 0.22 | Not recorded | 693115d6d6b0dd3342933d87 | | InferCept | 0.21 | Not recorded | 693115d6d6b0dd3342933d87 | | InferCept | 0.16 | Not recorded | 693115d6d6b0dd3342933d87 | | vLLM | 0.15 | Not recorded | 693115d6d6b0dd3342933d87 | | vLLM | 0.11 | Not recorded | 693115d6d6b0dd3342933d87 | ### AugServe Effective Throughput (req/s): 0.6. Result ID: 6996ce165698665ec41a0ffc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"2.0"}. This record does not assert independent reproduction. ### AugServe Effective Throughput (req/s): 0.57. Result ID: 6996ce165698665ec41a0ffd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"3.0"}. This record does not assert independent reproduction. ### AugServe Effective Throughput (req/s): 0.51. Result ID: 6996ce165698665ec41a0ffe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"4.0"}. This record does not assert independent reproduction. ### InferCept Effective Throughput (req/s): 0.27. Result ID: 6996ce165698665ec41a0ff9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"2.0"}. This record does not assert independent reproduction. ### vLLM Effective Throughput (req/s): 0.22. Result ID: 6996ce165698665ec41a0ff6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"2.0"}. This record does not assert independent reproduction. ### InferCept Effective Throughput (req/s): 0.21. Result ID: 6996ce165698665ec41a0ffa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"3.0"}. This record does not assert independent reproduction. ### InferCept Effective Throughput (req/s): 0.16. Result ID: 6996ce165698665ec41a0ffb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"4.0"}. This record does not assert independent reproduction. ### vLLM Effective Throughput (req/s): 0.15. Result ID: 6996ce165698665ec41a0ff7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"3.0"}. This record does not assert independent reproduction. ### vLLM Effective Throughput (req/s): 0.11. Result ID: 6996ce165698665ec41a0ff8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"OPT-13B","GPU":"H800","Request rate (req/s)":"4.0"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.