Model rankings
1319 models · 5 boards · 2 sources
Best images
Best open model
Best from China
Top three open models
Cosmos3-Super-Text2Image (agentic)NVIDIA · #8 overall1,227
HiDream-O1-Image-Dev-2604HiDream · #19 overall1,187
ERNIE ImageBaidu · #28 overall1,169
This week's changes
Reve 2.0Image · #2New
Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Intelligence · #5New
claude-sonnet-5-thinkingCoding · #5New
Fish Audio S2.1 ProSpeech · #15New
OpenAudio S1 MiniSpeech · #50New
| Rank | Change | Model | Score | 95% interval |
|---|---|---|---|---|
| 1 | — | Cosmos3-Super-Text2Image (agentic)OpenNVIDIA | 1,227 | ±9 |
| 2 | — | HiDream-O1-Image-Dev-2604OpenHiDream · 1 more variant | 1,187 | ±8 |
| 3 | ▲ 1 | ERNIE ImageOpenBaidu | 1,169 | ±10 |
| 4 | ▼ 1 | Ideogram 4.0 QualityOpenIdeogram | 1,169 | ±10 |
| 5 | — | ERNIE Image TurboOpenBaidu | 1,163 | ±9 |
| 6 | ▲ 1 | FLUX.2 [dev]OpenBlack Forest Labs | 1,156 | ±8 |
| 7 | ▲ 1 | Qwen Image Max 2512OpenAlibaba | 1,156 | ±9 |
| 8 | ▼ 2 | FLUX.2 [dev] TurboOpenFal | 1,156 | ±8 |
| 9 | — | FLUX.2 [dev] FlashOpenFal | 1,136 | ±8 |
| 10 | — | HunyuanImage 3.0 (Fal)OpenTencent | 1,125 | ±8 |
| 11 | — | FLUX.2 [klein] 9BOpenBlack Forest Labs | 1,121 | ±8 |
| 12 | — | HunyuanImage 3.0 Instruct (Fal)OpenTencent | 1,115 | ±8 |
| 13 | — | HiDream-O1-ImageOpenHiDream | 1,111 | ±9 |
| 14 | — | Z-Image TurboOpenAlibaba | 1,104 | ±8 |
| 15 | — | FLUX.2 [klein] Base 9BOpenBlack Forest Labs | 1,084 | ±8 |
| 16 | — | SRPOOpenTencent | 1,076 | ±8 |
| 17 | — | FIBOOpenBria | 1,071 | ±8 |
| 18 | — | HunyuanImage 2.1OpenTencent | 1,068 | ±8 |
| 19 | ▲ 1 | HiDream-I1-DevOpenHiDream | 1,065 | ±9 |
| 20 | ▼ 1 | FLUX.2 [klein] 4BOpenBlack Forest Labs | 1,062 | ±8 |
| 21 | — | Qwen ImageOpenAlibaba | 1,060 | ±8 |
| 22 | — | HiDream-I1-FastOpenHiDream | 1,055 | ±8 |
| 23 | — | GLM-ImageOpenZ AI | 1,047 | ±8 |
| 24 | — | Infinity 8BOpenByteDance | 1,045 | ±8 |
| 25 | ▲ 1 | FLUX.1 Krea [dev]OpenBlack Forest Labs | 1,031 | ±12 |
| 25 | — | LongCat ImageOpenMeituan | 1,031 | ±9 |
| 27 | — | FLUX.1 [dev]OpenBlack Forest Labs | 1,028 | ±8 |
| 28 | — | Stable Diffusion 3.5 LargeOpenStability.ai | 1,024 | ±8 |
| 29 | — | Stable Diffusion 3.5 Large TurboOpenStability.ai | 1,021 | ±8 |
| 30 | — | FLUX.1 [schnell]OpenBlack Forest Labs | 1,000 | ±0 |
| 31 | — | FLUX.2 [klein] Base 4BOpenBlack Forest Labs | 970 | ±9 |
| 32 | — | Lumina Image v2OpenOpenGVLab | 968 | ±8 |
| 33 | — | Playground v2.5OpenPlayground AI | 967 | ±8 |
| 34 | — | Stable Diffusion 3.5 MediumOpenStability.ai | 946 | ±9 |
| 35 | — | Sana Sprint 1.6BOpenNVIDIA | 932 | ±9 |
| 36 | — | Stable Diffusion 3 MediumOpenStability.ai | 911 | ±8 |
| 37 | — | SDXL LightningOpenByteDance | 906 | ±8 |
| 38 | — | OmniGen V2OpenVectorSpaceLab | 904 | ±9 |
| 39 | ▲ 1 | Bria 3.2OpenBria | 899 | ±10 |
| 40 | ▼ 1 | BagelOpenByteDance | 899 | ±10 |
| 41 | — | Stable Diffusion XL 1.0OpenStability.ai | 874 | ±9 |
| 42 | — | Stable Diffusion 2.1OpenStability.ai | 749 | ±11 |
| 43 | — | Janus ProOpenDeepSeek | 713 | ±12 |
| 44 | — | Stable Diffusion 1.5OpenStability.ai | 660 | ±13 |
Intelligence over time
Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi