# Retrieval-Augmented Generation on Medical (Cost Evaluation)

Source: [Retrieval-Augmented Generation on Medical (Cost Evaluation)](<https://www.sota2.com/research/sota/retrieval-augmented-generation-on-medical-cost-evaluation>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-18T07:52:25.385Z

Evaluation of the computational cost and latency associated with Retrieval-Augmented Generation systems using the Medical dataset, focusing on indexing time, querying time, and token usage costs.

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 | Retrieval-Augmented Generation |
| Dataset (including any stated split) | Medical |
| Metric | Indexing Time (minutes) |
| Unit | TIME |
| Direction | Lower is better |
| Results loaded | 11 of 11 |
| 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

- [Indexing Time (minutes)](<https://www.sota2.com/research/sota/retrieval-augmented-generation-on-medical-cost-evaluation.md?metric=Indexing%20Time%20(minutes)>): Lower is better; unit: TIME.

- [Querying Time (minutes)](<https://www.sota2.com/research/sota/retrieval-augmented-generation-on-medical-cost-evaluation.md?metric=Querying%20Time%20(minutes)>): Lower is better; unit: TIME.

- [Input Token Cost (per query)](<https://www.sota2.com/research/sota/retrieval-augmented-generation-on-medical-cost-evaluation.md?metric=Input%20Token%20Cost%20(per%20query)>): Lower is better; unit: SCALAR.

- [Output Token Cost (per query)](<https://www.sota2.com/research/sota/retrieval-augmented-generation-on-medical-cost-evaluation.md?metric=Output%20Token%20Cost%20(per%20query)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Indexing Time (minutes) | Result date | Report paper |
| --- | --- | --- | --- |
| NaiveRAG | 7 | Not recorded | 6912c3052828b7cc6937dbc7 |
| MIXRAG | 11 | Not recorded | 6912c3052828b7cc6937dbc7 |
| G-Retriever | 12 | Not recorded | 6912c3052828b7cc6937dbc7 |
| AGRAG | 16 | Not recorded | 6912c3052828b7cc6937dbc7 |
| L.LLMLingua | 37 | Not recorded | 6912c3052828b7cc6937dbc7 |
| SiReRAG | 40 | Not recorded | 6912c3052828b7cc6937dbc7 |
| HippoRAG2 | 46 | Not recorded | 6912c3052828b7cc6937dbc7 |
| F.GraphRAG | 48 | Not recorded | 6912c3052828b7cc6937dbc7 |
| LightRAG | 116 | Not recorded | 6912c3052828b7cc6937dbc7 |
| RAPTOR | 117 | Not recorded | 6912c3052828b7cc6937dbc7 |
| GraphRAG | 194 | Not recorded | 6912c3052828b7cc6937dbc7 |

### NaiveRAG

Indexing Time (minutes): 7. Result ID: 69ba59a047f34ce924a3f994.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72b51e209131cea5810f.md>)

Recorded evaluation configuration: {"Category":"Traditional RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### MIXRAG

Indexing Time (minutes): 11. Result ID: 69ba59a047f34ce924a3f999.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b911e209131cea892a9.md>)

Recorded evaluation configuration: {"Category":"Community-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### G-Retriever

Indexing Time (minutes): 12. Result ID: 69ba59a047f34ce924a3f998.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73541e209131cea5bdef.md>)

[Implementation](<https://github.com/XiaoxinHe/G-Retriever>)

Recorded evaluation configuration: {"Category":"Community-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### AGRAG

Indexing Time (minutes): 16. Result ID: 69ba59a047f34ce924a3f99e.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

Recorded evaluation configuration: {"Category":"Walking-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### L.LLMLingua

Indexing Time (minutes): 37. Result ID: 69ba59a047f34ce924a3f995.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74701e209131cea61c6d.md>)

[Implementation](<https://aka.ms/LLMLingua>)

Recorded evaluation configuration: {"Category":"Traditional RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### SiReRAG

Indexing Time (minutes): 40. Result ID: 69ba59a047f34ce924a3f997.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77d71e209131cea75814.md>)

Recorded evaluation configuration: {"Category":"Tree-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### HippoRAG2

Indexing Time (minutes): 46. Result ID: 69ba59a047f34ce924a3f99c.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79821e209131cea7de8a.md>)

Recorded evaluation configuration: {"Category":"Walking-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### F.GraphRAG

Indexing Time (minutes): 48. Result ID: 69ba59a047f34ce924a3f99d.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

Recorded evaluation configuration: {"Category":"Walking-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### LightRAG

Indexing Time (minutes): 116. Result ID: 69ba59a047f34ce924a3f99b.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78461e209131cea778bd.md>)

Recorded evaluation configuration: {"Category":"Community-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### RAPTOR

Indexing Time (minutes): 117. Result ID: 69ba59a047f34ce924a3f996.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71fe1e209131cea53fc2.md>)

Recorded evaluation configuration: {"Category":"Tree-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

### GraphRAG

Indexing Time (minutes): 194. Result ID: 69ba59a047f34ce924a3f99a.

[Reporting paper](<https://www.sota2.com/research/paper/6912c3052828b7cc6937dbc7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770d1e209131cea70c5a.md>)

Recorded evaluation configuration: {"Category":"Community-based RAG","LLM Backbone":"Qwen2.5-14B-Instruct"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
