# Image Classification on CLIP Zero-shot Evaluation Suite (10 datasets) Source: [Image Classification on CLIP Zero-shot Evaluation Suite (10 datasets)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:38:57.445Z Evaluation of Image Classification performance using a zero-shot protocol across 10 specific datasets included in the CLIP evaluation suite, reporting individual scores for Food101, CIFAR10, CIFAR100, SUN397, Cars, Aircraft, DTD, Pets, Cal101, and Flowers, along with the average. 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 | | Dataset (including any stated split) | CLIP Zero-shot Evaluation Suite (10 datasets) | | Metric | Cars Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [Cars Accuracy](): Higher is better; unit: PERCENT. - [Food101 Accuracy](): Higher is better; unit: PERCENT. - [CIFAR10 Accuracy](): Higher is better; unit: PERCENT. - [CIFAR100 Accuracy](): Higher is better; unit: PERCENT. - [SUN397 Accuracy](): Higher is better; unit: PERCENT. - [Aircraft Accuracy](): Higher is better; unit: PERCENT. - [DTD Accuracy](): Higher is better; unit: PERCENT. - [Pets Accuracy](): Higher is better; unit: PERCENT. - [Cal101 Accuracy](): Higher is better; unit: PERCENT. - [Flowers Accuracy](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Cars Accuracy | Result date | Report paper | | --- | --- | --- | --- | | CLIP-L | 89.6 | Not recorded | 68fa78231e209131cea76f2e | | CLIP-B | 83.5 | Not recorded | 68fa78231e209131cea76f2e | | SAIL-L-NV2 | 44.6 | Not recorded | 68fa78231e209131cea76f2e | | SAIL-L-NV2 | 42.9 | Not recorded | 68fa78231e209131cea76f2e | | SAIL-B-NV2 | 35.8 | Not recorded | 68fa78231e209131cea76f2e | | SAIL-L-GTE | 33 | Not recorded | 68fa78231e209131cea76f2e | | SAIL-L-GTE | 32 | Not recorded | 68fa78231e209131cea76f2e | | DreamLIP | 29.7 | Not recorded | 68fa78231e209131cea76f2e | | SAIL-B-GTE | 28.1 | Not recorded | 68fa78231e209131cea76f2e | | CLIP-B | 14.7 | Not recorded | 68fa78231e209131cea76f2e | | LiT | 13.2 | Not recorded | 68fa78231e209131cea76f2e | | ShareLock+ | 11.5 | Not recorded | 68fa78231e209131cea76f2e | | SAIL-B-GTE | 9.2 | Not recorded | 68fa78231e209131cea76f2e | | LiT | 3 | Not recorded | 68fa78231e209131cea76f2e | | DreamLIP | 2.8 | Not recorded | 68fa78231e209131cea76f2e | | CLIP-B | 1.2 | Not recorded | 68fa78231e209131cea76f2e | ### CLIP-L Cars Accuracy: 89.6. Result ID: 69957de33badfaffe6757497. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"LAION400M","Model Architecture":"ViT-L/14","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### CLIP-B Cars Accuracy: 83.5. Result ID: 69957de33badfaffe6757492. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"LAION400M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### SAIL-L-NV2 Cars Accuracy: 44.6. Result ID: 69957de33badfaffe6757496. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"23M Merged","Model Architecture":"ViT-L/14","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### SAIL-L-NV2 Cars Accuracy: 42.9. Result ID: 69957de33badfaffe6757495. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"CC12M","Model Architecture":"ViT-L/14","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### SAIL-B-NV2 Cars Accuracy: 35.8. Result ID: 69957de33badfaffe6757491. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"CC12M","Model Architecture":"ViT-B/16","Patch Size":14,"Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### SAIL-L-GTE Cars Accuracy: 33. Result ID: 69957de33badfaffe6757493. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"CC12M","Model Architecture":"ViT-L/14","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### SAIL-L-GTE Cars Accuracy: 32. Result ID: 69957de33badfaffe6757494. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"23M Merged","Model Architecture":"ViT-L/14","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### DreamLIP Cars Accuracy: 29.7. Result ID: 69957de33badfaffe675748d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"CC12M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### SAIL-B-GTE Cars Accuracy: 28.1. Result ID: 69957de33badfaffe6757490. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"CC12M","Model Architecture":"ViT-B/16","Patch Size":14,"Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### CLIP-B Cars Accuracy: 14.7. Result ID: 69957de33badfaffe675748c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"CC12M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### LiT Cars Accuracy: 13.2. Result ID: 69957de33badfaffe675748e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"CC12M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### ShareLock+ Cars Accuracy: 11.5. Result ID: 69957de33badfaffe675748f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"CC12M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### SAIL-B-GTE Cars Accuracy: 9.2. Result ID: 69957de33badfaffe675748b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"CC3M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### LiT Cars Accuracy: 3. Result ID: 69957de33badfaffe675748a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"CC3M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### DreamLIP Cars Accuracy: 2.8. Result ID: 69957de33badfaffe6757489. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"CC3M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### CLIP-B Cars Accuracy: 1.2. Result ID: 69957de33badfaffe6757488. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"CC3M","Model Architecture":"ViT-B/16","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.