# Financial NLP on FinGPT

Source: [Financial NLP on FinGPT](<https://www.sota2.com/research/sota/financial-nlp-on-fingpt>)

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

Data updated: 2026-02-26T05:30:57.245Z

Evaluation of general Financial Natural Language Processing capabilities on the FinGPT dataset, measured using accuracy.

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 | Financial NLP |
| Dataset (including any stated split) | FinGPT |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 28 of 28 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/financial-nlp-on-fingpt.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Dense Model | 86.9 | Not recorded | 6979984ba2b28e761d87cd0d |
| FT(Llama3.2) | 86.7 | Not recorded | 6979984ba2b28e761d87cd0d |
| GradPruner | 86.7 | Not recorded | 6979984ba2b28e761d87cd0d |
| Dense Model | 86.2 | Not recorded | 6979984ba2b28e761d87cd0d |
| Dense Model | 85.3 | Not recorded | 6979984ba2b28e761d87cd0d |
| GradPruner | 84.6 | Not recorded | 6979984ba2b28e761d87cd0d |
| GradPruner | 84 | Not recorded | 6979984ba2b28e761d87cd0d |
| SAT | 83.2 | Not recorded | 6979984ba2b28e761d87cd0d |
| Dense Model | 83.1 | Not recorded | 6979984ba2b28e761d87cd0d |
| FT(Llama3.2) | 82.5 | Not recorded | 6979984ba2b28e761d87cd0d |
| MINITRON | 82.3 | Not recorded | 6979984ba2b28e761d87cd0d |
| SAT | 82.3 | Not recorded | 6979984ba2b28e761d87cd0d |
| Laco | 81.9 | Not recorded | 6979984ba2b28e761d87cd0d |
| LLMPruner | 81.8 | Not recorded | 6979984ba2b28e761d87cd0d |
| GradPruner | 81.7 | Not recorded | 6979984ba2b28e761d87cd0d |
| LLMPruner | 81.7 | Not recorded | 6979984ba2b28e761d87cd0d |
| Laco | 81.4 | Not recorded | 6979984ba2b28e761d87cd0d |
| APT | 81.3 | Not recorded | 6979984ba2b28e761d87cd0d |
| MINITRON | 80.8 | Not recorded | 6979984ba2b28e761d87cd0d |
| SAT | 80.7 | Not recorded | 6979984ba2b28e761d87cd0d |
| Laco | 80.4 | Not recorded | 6979984ba2b28e761d87cd0d |
| APT | 80.2 | Not recorded | 6979984ba2b28e761d87cd0d |
| MINITRON | 80.2 | Not recorded | 6979984ba2b28e761d87cd0d |
| LLMPruner | 79.3 | Not recorded | 6979984ba2b28e761d87cd0d |
| SAT | 77.9 | Not recorded | 6979984ba2b28e761d87cd0d |
| Laco | 77.5 | Not recorded | 6979984ba2b28e761d87cd0d |
| LLMPruner | 76.6 | Not recorded | 6979984ba2b28e761d87cd0d |
| MINITRON | 76.5 | Not recorded | 6979984ba2b28e761d87cd0d |

### Dense Model

Accuracy: 86.9. Result ID: 699668e05698665ec4af6c71.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"FFT","Sparsity":"0%"}. This record does not assert independent reproduction.

### FT(Llama3.2)

Accuracy: 86.7. Result ID: 699668e05698665ec4af6c76.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Base Model":"Llama3.2-3B","Fine-tuning Protocol":"FFT","Sparsity":"0%"}. This record does not assert independent reproduction.

### GradPruner

Accuracy: 86.7. Result ID: 699668e05698665ec4af6c77.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### Dense Model

Accuracy: 86.2. Result ID: 699668e05698665ec4af6c80.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"FFT","Sparsity":"0%"}. This record does not assert independent reproduction.

### Dense Model

Accuracy: 85.3. Result ID: 699668e05698665ec4af6c86.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"LoRA","Sparsity":"0%"}. This record does not assert independent reproduction.

### GradPruner

Accuracy: 84.6. Result ID: 699668e05698665ec4af6c85.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### GradPruner

Accuracy: 84. Result ID: 699668e05698665ec4af6c8c.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### SAT

Accuracy: 83.2. Result ID: 699668e05698665ec4af6c75.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d20b.md>)

[Implementation](<https://github.com/OpenDFM/SAT>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### Dense Model

Accuracy: 83.1. Result ID: 699668e05698665ec4af6c78.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"LoRA","Sparsity":"0%"}. This record does not assert independent reproduction.

### FT(Llama3.2)

Accuracy: 82.5. Result ID: 699668e05698665ec4af6c7e.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Base Model":"Llama3.2-3B","Fine-tuning Protocol":"LoRA","Sparsity":"0%"}. This record does not assert independent reproduction.

### MINITRON

Accuracy: 82.3. Result ID: 699668e05698665ec4af6c83.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75901e209131cea68abe.md>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### SAT

Accuracy: 82.3. Result ID: 699668e05698665ec4af6c84.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d20b.md>)

[Implementation](<https://github.com/OpenDFM/SAT>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### Laco

Accuracy: 81.9. Result ID: 699668e05698665ec4af6c82.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75321e209131cea66502.md>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### LLMPruner

Accuracy: 81.8. Result ID: 699668e05698665ec4af6c72.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70af1e209131cea4c8b2.md>)

[Implementation](<https://github.com/horseee/LLM-Pruner>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### GradPruner

Accuracy: 81.7. Result ID: 699668e05698665ec4af6c7f.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### LLMPruner

Accuracy: 81.7. Result ID: 699668e05698665ec4af6c81.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70af1e209131cea4c8b2.md>)

[Implementation](<https://github.com/horseee/LLM-Pruner>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### Laco

Accuracy: 81.4. Result ID: 699668e05698665ec4af6c73.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75321e209131cea66502.md>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### APT

Accuracy: 81.3. Result ID: 699668e05698665ec4af6c8b.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73771e209131cea5cb42.md>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### MINITRON

Accuracy: 80.8. Result ID: 699668e05698665ec4af6c74.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75901e209131cea68abe.md>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"FFT","Sparsity":"40%"}. This record does not assert independent reproduction.

### SAT

Accuracy: 80.7. Result ID: 699668e05698665ec4af6c8a.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d20b.md>)

[Implementation](<https://github.com/OpenDFM/SAT>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### Laco

Accuracy: 80.4. Result ID: 699668e05698665ec4af6c88.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75321e209131cea66502.md>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### APT

Accuracy: 80.2. Result ID: 699668e05698665ec4af6c7d.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73771e209131cea5cb42.md>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### MINITRON

Accuracy: 80.2. Result ID: 699668e05698665ec4af6c89.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75901e209131cea68abe.md>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### LLMPruner

Accuracy: 79.3. Result ID: 699668e05698665ec4af6c87.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70af1e209131cea4c8b2.md>)

[Implementation](<https://github.com/horseee/LLM-Pruner>)

Recorded evaluation configuration: {"Base Model":"Mistral-7B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### SAT

Accuracy: 77.9. Result ID: 699668e05698665ec4af6c7c.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d20b.md>)

[Implementation](<https://github.com/OpenDFM/SAT>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### Laco

Accuracy: 77.5. Result ID: 699668e05698665ec4af6c7a.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75321e209131cea66502.md>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### LLMPruner

Accuracy: 76.6. Result ID: 699668e05698665ec4af6c79.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70af1e209131cea4c8b2.md>)

[Implementation](<https://github.com/horseee/LLM-Pruner>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

### MINITRON

Accuracy: 76.5. Result ID: 699668e05698665ec4af6c7b.

[Reporting paper](<https://www.sota2.com/research/paper/6979984ba2b28e761d87cd0d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75901e209131cea68abe.md>)

Recorded evaluation configuration: {"Base Model":"Llama3.1-8B","Fine-tuning Protocol":"LoRA","Sparsity":"40%"}. This record does not assert independent reproduction.

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