# Image Classification Inference on 1 Million (1m classes)

Source: [Image Classification Inference on 1 Million (1m classes)](<https://www.sota2.com/research/sota/image-classification-inference-on-1-million-1m-classes>)

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

Data updated: 2026-04-02T07:54:54.597Z

Evaluation of model inference performance on the 1 Million (1m classes) dataset, measured using specific inference metrics (td, tf, tc, tn, Ks, Kt).

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 | Image Classification Inference |
| Dataset (including any stated split) | 1 Million (1m classes) |
| Metric | Inference Time Component (td) |
| Unit | TIME |
| Direction | Lower is better |
| Results loaded | 2 of 2 |
| 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

- [Inference Time Component (td)](<https://www.sota2.com/research/sota/image-classification-inference-on-1-million-1m-classes.md?metric=Inference%20Time%20Component%20(td)>): Lower is better; unit: TIME.

- [Inference Time Component (tf)](<https://www.sota2.com/research/sota/image-classification-inference-on-1-million-1m-classes.md?metric=Inference%20Time%20Component%20(tf)>): Lower is better; unit: TIME.

- [Inference Time Component (tc)](<https://www.sota2.com/research/sota/image-classification-inference-on-1-million-1m-classes.md?metric=Inference%20Time%20Component%20(tc)>): Lower is better; unit: TIME.

- [Inference Metric (tn)](<https://www.sota2.com/research/sota/image-classification-inference-on-1-million-1m-classes.md?metric=Inference%20Metric%20(tn)>): Higher is better; unit: SCALAR.

- [K Score (Ks)](<https://www.sota2.com/research/sota/image-classification-inference-on-1-million-1m-classes.md?metric=K%20Score%20(Ks)>): Higher is better; unit: SCALAR.

- [K Factor (Kt)](<https://www.sota2.com/research/sota/image-classification-inference-on-1-million-1m-classes.md?metric=K%20Factor%20(Kt)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Inference Time Component (td) | Result date | Report paper |
| --- | --- | --- | --- |
| Proposed Method | 4325 | Not recorded | 69cdf80e0636e5bf4afcca74 |
| Proposed Method | 4536 | Not recorded | 69cdf80e0636e5bf4afcca74 |

### Proposed Method

Inference Time Component (td): 4325. Result ID: 69ce20bb47f34ce924b72b62.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf80e0636e5bf4afcca74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf80e0636e5bf4afcca74.md>)

Recorded evaluation configuration: {"ndim":"82","Batch size":"1024"}. This record does not assert independent reproduction.

### Proposed Method

Inference Time Component (td): 4536. Result ID: 69ce20bb47f34ce924b72b60.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf80e0636e5bf4afcca74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf80e0636e5bf4afcca74.md>)

Recorded evaluation configuration: {"ndim":"47","Batch size":"3584"}. This record does not assert independent reproduction.

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