Benchmark
VLMsAreBlind Benchmark
What is VLMsAreBlind?
A vision-language benchmark that probes blind spots and brittle reasoning in multimodal models. Benchmark Radar tracks 4 reported scores for VLMsAreBlind…
- Released
- 2024-07-09
- Modality
- multimodal
- Evaluates
- multimodal, reasoning, vision
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 4
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/vlmsareblind?top_n=500
VLMsAreBlind results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Qwen3.5-122B-A10B | Qwen | 0.967 | 2026-02-24 | evidence |
| Qwen3.5-27B | Qwen | 0.969 | 2026-02-24 | evidence |
| Qwen3.5-35B-A3B | Qwen | 0.97 | 2026-02-24 | evidence |
| Qwen3.6-27B | Qwen | 0.97 | 2026-04-21 | 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.