Benchmark
BLINK
BLINK: Multimodal Large Language Models Can See but Not Perceive. A benchmark for multimodal language models focusing on core visual perception abilities…
- Modality
- multimodal
- Categories
- multimodal, reasoning, spatial_reasoning, 3d, vision
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 15
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| LFM2.5-VL-3B | Liquid AI | 0.615 | 2026-08-12 | evidence |
| North Micro Vision Instruct | Cohere | 0.527 | 2026-08-12 | evidence |
| Phi-4-multimodal-instruct | Microsoft | 0.613 | 2025-02-01 | evidence |
| Qwen3 VL 235B A22B Instruct | Qwen | 0.707 | 2025-09-22 | evidence |
| Qwen3 VL 235B A22B Thinking | Qwen | 0.671 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Instruct | Qwen | 0.677 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Thinking | Qwen | 0.654 | 2025-09-22 | evidence |
| Qwen3 VL 32B Instruct | Qwen | 0.673 | 2025-09-22 | evidence |
| Qwen3 VL 32B Thinking | Qwen | 0.685 | 2025-09-22 | evidence |
| Qwen3 VL 4B Instruct | Qwen | 0.658 | 2025-09-22 | evidence |
| Qwen3 VL 4B Thinking | Qwen | 0.634 | 2025-09-22 | evidence |
| Qwen3 VL 8B Instruct | Qwen | 0.691 | 2025-09-22 | evidence |
| Qwen3 VL 8B Thinking | Qwen | 0.687 | 2025-09-22 | evidence |
| Seed 2.1 Pro | ByteDance | 0.814 | 2026-06-24 | evidence |
| Seed 2.1 Turbo | ByteDance | 0.794 | 2026-06-24 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.