CNSL-bench - Text Descriptions: leaderboard
CNSL-bench (2026) evaluates multimodal LLMs on Chinese National Sign Language: each entry of the standardized National Common Sign Language Dictionary is posed as a four-way multiple-choice question from a single input modality. This board is accuracy in percent identifying the sign from the dictionary's textual description of its articulation; higher is better, and chance is 25.
Metric: Accuracy (%) identifying the sign's meaning from the dictionary's textual description of its articulation, four-way multiple choice over the 6,707 sign entries of the National Common Sign Language Dictionary (distractors sampled at random), official sampling settings per model, accuracy by exact match of the chosen option; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 19 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-5 | 89.64 | #178 | |
| 2 | Gemini 2.5 Pro | Medium | 84.79 | #276 |
| 3 | Gemini 2.5 Flash | 79.95 | #463 | |
| 4 | Qwen 3 VL Plus | 76.68 | #512 | |
| 5 | Qwen-VL-Max | 72.62 | #1400 | |
| 6 | Qwen-VL-Plus | 72.41 | #5818 | |
| 7 | Qwen 3 VL 8B (Thinking) | 70.61 | #910 | |
| 8 | GPT-4o | 69.03 | #751 | |
| 9 | GLM-4.1V-9B (Thinking) | 68.24 | ||
| 10 | GPT-4o Mini | 67.57 | #1667 | |
| 11 | InternVL3.5-8B | 67.53 | #1604 | |
| 12 | Qwen 3 VL 8B Instruct | 67.06 | #1029 | |
| 13 | Qwen 3 VL 2B Instruct | 62.83 | #2367 | |
| 14 | Qwen 3 VL 2B (Thinking) | 61.29 | #2535 | |
| 15 | Qwen 2.5 VL 3B Instruct | 60.07 | #2713 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=cnsl-bench-text-descriptions · How It Works · Data refreshed daily, snapshot 2026-11-21.