# Agentic Workflow Performance (Static) on FinanceBench + CrewAI

Source: [Agentic Workflow Performance (Static) on FinanceBench + CrewAI](<https://www.sota2.com/research/sota/agentic-workflow-performance-static-on-financebench-plus-crewai>)

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

Data updated: 2026-05-08T11:17:08.298Z

Evaluation of static agentic workflow performance using the FinanceBench + CrewAI dataset, measured by latency, time-to-first-byte (TTFT), and hit rate.

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 | Agentic Workflow Performance (Static) |
| Dataset (including any stated split) | FinanceBench + CrewAI |
| Metric | Latency (s) |
| Unit | TIME |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| 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

- [Latency (s)](<https://www.sota2.com/research/sota/agentic-workflow-performance-static-on-financebench-plus-crewai.md?metric=Latency%20(s)>): Lower is better; unit: TIME.

- [TTFT (s)](<https://www.sota2.com/research/sota/agentic-workflow-performance-static-on-financebench-plus-crewai.md?metric=TTFT%20(s)>): Lower is better; unit: TIME.

- [Hit Rate](<https://www.sota2.com/research/sota/agentic-workflow-performance-static-on-financebench-plus-crewai.md?metric=Hit%20Rate>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Latency (s) | Result date | Report paper |
| --- | --- | --- | --- |
| Full PBKV | 65.01 | Not recorded | 69fd6e5c5497b3b670c89726 |
| KVFlow | 80.13 | Not recorded | 69fd6e5c5497b3b670c89726 |
| Full PBKV | 80.53 | Not recorded | 69fd6e5c5497b3b670c89726 |
| LRU | 96.11 | Not recorded | 69fd6e5c5497b3b670c89726 |
| KVFlow | 101.57 | Not recorded | 69fd6e5c5497b3b670c89726 |
| LRU | 130.34 | Not recorded | 69fd6e5c5497b3b670c89726 |

### Full PBKV

Latency (s): 65.01. Result ID: 69fdc62747f34ce924e1a953.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5c5497b3b670c89726.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e5c5497b3b670c89726.md>)

Recorded evaluation configuration: {"Model":"Qwen3-14B"}. This record does not assert independent reproduction.

### KVFlow

Latency (s): 80.13. Result ID: 69fdc62747f34ce924e1a952.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5c5497b3b670c89726.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ed1e209131cea80099.md>)

Recorded evaluation configuration: {"Model":"Qwen3-14B"}. This record does not assert independent reproduction.

### Full PBKV

Latency (s): 80.53. Result ID: 69fdc62747f34ce924e1a950.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5c5497b3b670c89726.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e5c5497b3b670c89726.md>)

Recorded evaluation configuration: {"Model":"Qwen3-32B"}. This record does not assert independent reproduction.

### LRU

Latency (s): 96.11. Result ID: 69fdc62747f34ce924e1a951.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5c5497b3b670c89726.md>)

Recorded evaluation configuration: {"Model":"Qwen3-14B"}. This record does not assert independent reproduction.

### KVFlow

Latency (s): 101.57. Result ID: 69fdc62747f34ce924e1a94f.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5c5497b3b670c89726.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ed1e209131cea80099.md>)

Recorded evaluation configuration: {"Model":"Qwen3-32B"}. This record does not assert independent reproduction.

### LRU

Latency (s): 130.34. Result ID: 69fdc62747f34ce924e1a94e.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e5c5497b3b670c89726.md>)

Recorded evaluation configuration: {"Model":"Qwen3-32B"}. This record does not assert independent reproduction.

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