# Text-to-interleaved output generation on UEval

Source: [Text-to-interleaved output generation on UEval](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval>)

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

Data updated: 2026-06-12T08:31:10.337Z

Evaluation of models generating interleaved text outputs on the UEval dataset, assessing performance across categories such as Space, Textbook, Diagram, Paper, Art, Life, Tech, and Exercise.

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 | Text-to-interleaved output generation |
| Dataset (including any stated split) | UEval |
| Metric | Accuracy (Space) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 13 of 13 |
| 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 (Space)](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Accuracy%20(Space)>): Higher is better; unit: PERCENT.

- [Accuracy (Textbook)](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Accuracy%20(Textbook)>): Higher is better; unit: PERCENT.

- [Accuracy (Diagram)](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Accuracy%20(Diagram)>): Higher is better; unit: PERCENT.

- [Accuracy (Paper)](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Accuracy%20(Paper)>): Higher is better; unit: PERCENT.

- [Accuracy (Art)](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Accuracy%20(Art)>): Higher is better; unit: PERCENT.

- [Accuracy (Life)](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Accuracy%20(Life)>): Higher is better; unit: PERCENT.

- [Accuracy (Tech)](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Accuracy%20(Tech)>): Higher is better; unit: PERCENT.

- [Accuracy (Exercise)](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Accuracy%20(Exercise)>): Higher is better; unit: PERCENT.

- [Average Accuracy](<https://www.sota2.com/research/sota/text-to-interleaved-output-generation-on-ueval.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Space) | Result date | Report paper |
| --- | --- | --- | --- |
| Reference | 96.2 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| GPT-5-Thinking | 84 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| Nano Banana Pro | 79.4 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| Nano Banana | 78 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| GPT-5-Instant | 77.3 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| InterleaveThinker | 65.8 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| Gemini-2.0-Flash | 65.2 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| InterleaveThinker | 62.1 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| Emu3.5 | 59.1 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| BAGEL | 29.8 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| Show-o2 | 25.4 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| Janus-Pro | 21 | Not recorded | 6a2b92d9d33dfb23e783b32e |
| MMaDA | 10.8 | Not recorded | 6a2b92d9d33dfb23e783b32e |

### Reference

Accuracy (Space): 96.2. Result ID: 6a2bc3b7fbc373592c2a40e7.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

Recorded evaluation configuration: {"Category":"Reference"}. This record does not assert independent reproduction.

### GPT-5-Thinking

Accuracy (Space): 84. Result ID: 6a2bc3b7fbc373592c2a40ea.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

Recorded evaluation configuration: {"Category":"Proprietary Frontier Models"}. This record does not assert independent reproduction.

### Nano Banana Pro

Accuracy (Space): 79.4. Result ID: 6a2bc3b7fbc373592c2a40ec.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

Recorded evaluation configuration: {"Category":"Proprietary Frontier Models"}. This record does not assert independent reproduction.

### Nano Banana

Accuracy (Space): 78. Result ID: 6a2bc3b7fbc373592c2a40eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

Recorded evaluation configuration: {"Category":"Proprietary Frontier Models"}. This record does not assert independent reproduction.

### GPT-5-Instant

Accuracy (Space): 77.3. Result ID: 6a2bc3b7fbc373592c2a40e9.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

Recorded evaluation configuration: {"Category":"Proprietary Frontier Models"}. This record does not assert independent reproduction.

### InterleaveThinker

Accuracy (Space): 65.8. Result ID: 6a2bc3b7fbc373592c2a40f3.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Implementation](<https://github.com/zhengdian1/InterleaveThinker>)

Recorded evaluation configuration: {"Base Model":"Qwen-Image-Edit","Category":"Open-Sourced Models"}. This record does not assert independent reproduction.

### Gemini-2.0-Flash

Accuracy (Space): 65.2. Result ID: 6a2bc3b7fbc373592c2a40e8.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

Recorded evaluation configuration: {"Category":"Proprietary Frontier Models"}. This record does not assert independent reproduction.

### InterleaveThinker

Accuracy (Space): 62.1. Result ID: 6a2bc3b7fbc373592c2a40f2.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Implementation](<https://github.com/zhengdian1/InterleaveThinker>)

Recorded evaluation configuration: {"Base Model":"FLUX.2-klein-9B","Category":"Open-Sourced Models"}. This record does not assert independent reproduction.

### Emu3.5

Accuracy (Space): 59.1. Result ID: 6a2bc3b7fbc373592c2a40f1.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/690442653ef7ccd74e5e497b.md>)

Recorded evaluation configuration: {"Category":"Open-Sourced Models"}. This record does not assert independent reproduction.

### BAGEL

Accuracy (Space): 29.8. Result ID: 6a2bc3b7fbc373592c2a40f0.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a311e209131cea81aad.md>)

[Implementation](<https://bagel-ai.org/>)

Recorded evaluation configuration: {"Category":"Open-Sourced Models"}. This record does not assert independent reproduction.

### Show-o2

Accuracy (Space): 25.4. Result ID: 6a2bc3b7fbc373592c2a40ee.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b761e209131cea88a1f.md>)

Recorded evaluation configuration: {"Category":"Open-Sourced Models"}. This record does not assert independent reproduction.

### Janus-Pro

Accuracy (Space): 21. Result ID: 6a2bc3b7fbc373592c2a40ed.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76cc1e209131cea6f20a.md>)

[Implementation](<https://github.com/deepseek-ai/Janus>)

Recorded evaluation configuration: {"Category":"Open-Sourced Models"}. This record does not assert independent reproduction.

### MMaDA

Accuracy (Space): 10.8. Result ID: 6a2bc3b7fbc373592c2a40ef.

[Reporting paper](<https://www.sota2.com/research/paper/6a2b92d9d33dfb23e783b32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b951e209131cea8947a.md>)

Recorded evaluation configuration: {"Category":"Open-Sourced Models"}. This record does not assert independent reproduction.

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